chore(codex): trim filename comment to the non-obvious bit

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

.claude/skills/add-atomic-chat-tool/SKILL.md

@@ -0,0 +1,243 @@
+---
+name: add-atomic-chat-tool
+description: Add Atomic Chat MCP server so the container agent can call local models served by the Atomic Chat desktop app via its OpenAI-compatible API.
+---
+
+# Add Atomic Chat Integration
+
+This skill adds a stdio-based MCP server that exposes models running in the local [Atomic Chat](https://github.com/AtomicBot-ai/Atomic-Chat) desktop app as tools for the container agent. Claude remains the orchestrator but can offload work to local models served by Atomic Chat on `http://127.0.0.1:1337/v1` (OpenAI-compatible).
+
+Tools exposed:
+- `atomic_chat_list_models` — list models currently available in Atomic Chat (`GET /v1/models`)
+- `atomic_chat_generate` — send a prompt to a specified model and return the response (`POST /v1/chat/completions`)
+
+Model management (download, delete) is done through the **Atomic Chat desktop UI** — the app is a fork of Jan and manages its own model library.
+
+The skill ships the MCP server source in this folder and copies it into the agent-runner tree at install time, then wires it up with small edits to `index.ts`, `providers/claude.ts`, and `container-runner.ts`. No branch merge — all edits are additive and idempotent.
+
+## Phase 1: Pre-flight
+
+### Check if already applied
+
+Check if `container/agent-runner/src/atomic-chat-mcp-stdio.ts` exists. If it does, skip to Phase 3 (Configure).
+
+### Check prerequisites
+
+Verify Atomic Chat is installed and its local API server is running. On the host:
+
+```bash
+curl -s http://127.0.0.1:1337/v1/models | head
+```
+
+If the request fails:
+
+1. Install Atomic Chat from the [latest release](https://github.com/AtomicBot-ai/Atomic-Chat/releases) (macOS only for now — `atomic-chat.dmg`).
+2. Open the app.
+3. Open **Settings → Local API Server** and make sure it's enabled on port `1337`.
+4. Go to the **Hub** (or **Models**) tab and download at least one model (e.g. Llama 3.2 3B, Qwen 2.5 Coder 7B).
+5. Load the model once by sending any message in Atomic Chat's UI to warm it up.
+
+## Phase 2: Apply Code Changes
+
+### Copy the MCP server source
+
+```bash
+cp .claude/skills/add-atomic-chat-tool/atomic-chat-mcp-stdio.ts container/agent-runner/src/atomic-chat-mcp-stdio.ts
+```
+
+### Register the MCP server in the agent-runner
+
+Edit `container/agent-runner/src/index.ts`. Find the `mcpServers` object that currently looks like this:
+
+```ts
+  const mcpServers: Record<string, { command: string; args: string[]; env: Record<string, string> }> = {
+    nanoclaw: {
+      command: 'bun',
+      args: ['run', mcpServerPath],
+      env: {},
+    },
+  };
+```
+
+Add an `atomic_chat` entry alongside `nanoclaw`:
+
+```ts
+  const mcpServers: Record<string, { command: string; args: string[]; env: Record<string, string> }> = {
+    nanoclaw: {
+      command: 'bun',
+      args: ['run', mcpServerPath],
+      env: {},
+    },
+    atomic_chat: {
+      command: 'bun',
+      args: ['run', path.join(__dirname, 'atomic-chat-mcp-stdio.ts')],
+      env: {
+        ...(process.env.ATOMIC_CHAT_HOST ? { ATOMIC_CHAT_HOST: process.env.ATOMIC_CHAT_HOST } : {}),
+        ...(process.env.ATOMIC_CHAT_API_KEY ? { ATOMIC_CHAT_API_KEY: process.env.ATOMIC_CHAT_API_KEY } : {}),
+      },
+    },
+  };
+```
+
+### Add the tool glob to the allowlist
+
+Edit `container/agent-runner/src/providers/claude.ts`. Find `'mcp__nanoclaw__*',` in the `TOOL_ALLOWLIST` array and add `'mcp__atomic_chat__*',` on the following line:
+
+```ts
+  'mcp__nanoclaw__*',
+  'mcp__atomic_chat__*',
+];
+```
+
+### Forward host env vars into the container
+
+Edit `src/container-runner.ts` in `buildContainerArgs`. Find the `TZ` env line:
+
+```ts
+  args.push('-e', `TZ=${TIMEZONE}`);
+```
+
+Add ATOMIC_CHAT forwarding right after it:
+
+```ts
+  args.push('-e', `TZ=${TIMEZONE}`);
+
+  // Atomic Chat MCP tool: forward host overrides if set (default is host.docker.internal:1337).
+  if (process.env.ATOMIC_CHAT_HOST) {
+    args.push('-e', `ATOMIC_CHAT_HOST=${process.env.ATOMIC_CHAT_HOST}`);
+  }
+  if (process.env.ATOMIC_CHAT_API_KEY) {
+    args.push('-e', `ATOMIC_CHAT_API_KEY=${process.env.ATOMIC_CHAT_API_KEY}`);
+  }
+```
+
+### Surface `[ATOMIC]` log lines at info level
+
+In the same file, find the stderr logger:
+
+```ts
+  container.stderr?.on('data', (data) => {
+    for (const line of data.toString().trim().split('\n')) {
+      if (line) log.debug(line, { container: agentGroup.folder });
+    }
+  });
+```
+
+Replace it with:
+
+```ts
+  container.stderr?.on('data', (data) => {
+    for (const line of data.toString().trim().split('\n')) {
+      if (!line) continue;
+      if (line.includes('[ATOMIC]')) {
+        log.info(line, { container: agentGroup.folder });
+      } else {
+        log.debug(line, { container: agentGroup.folder });
+      }
+    }
+  });
+```
+
+### Add env-var stubs to `.env.example`
+
+Append to `.env.example`:
+
+```bash
+# Atomic Chat MCP tool (.claude/skills/add-atomic-chat-tool)
+# Override the host where Atomic Chat exposes its OpenAI-compatible API.
+# Default: http://host.docker.internal:1337 (with fallback to localhost)
+# ATOMIC_CHAT_HOST=http://host.docker.internal:1337
+
+# Optional API key. Leave unset for a local Atomic Chat install — it does not require auth.
+# ATOMIC_CHAT_API_KEY=
+```
+
+### Validate code changes
+
+```bash
+pnpm run build
+pnpm exec tsc -p container/agent-runner/tsconfig.json --noEmit
+./container/build.sh
+```
+
+All three must be clean before proceeding.
+
+## Phase 3: Configure
+
+### Set Atomic Chat host (optional)
+
+By default, the MCP server connects to `http://host.docker.internal:1337` (Docker Desktop) with a fallback to `localhost`. To use a custom host, add to `.env`:
+
+```bash
+ATOMIC_CHAT_HOST=http://your-atomic-chat-host:1337
+```
+
+### Set API key (optional)
+
+Atomic Chat does **not require authentication** when running locally — leave this unset. Only set it if you've put Atomic Chat behind a reverse proxy that enforces auth:
+
+```bash
+ATOMIC_CHAT_API_KEY=sk-...
+```
+
+### Restart the service
+
+```bash
+launchctl kickstart -k gui/$(id -u)/com.nanoclaw  # macOS
+# Linux: systemctl --user restart nanoclaw
+```
+
+## Phase 4: Verify
+
+### Test inference
+
+Tell the user:
+
+> Send a message like: "use atomic chat to tell me the capital of France"
+>
+> The agent should use `atomic_chat_list_models` to find available models, then `atomic_chat_generate` to get a response.
+
+### Check logs if needed
+
+```bash
+tail -f logs/nanoclaw.log | grep -i atomic
+```
+
+Look for:
+- `[ATOMIC] Listing models...` — list request started
+- `[ATOMIC] Found N models` — models discovered
+- `[ATOMIC] >>> Generating with <model>` — generation started
+- `[ATOMIC] <<< Done: <model> | Xs | N tokens | M chars` — generation completed
+
+## Troubleshooting
+
+### Agent says "Atomic Chat is not installed" or tries to run a CLI
+
+The agent is looking for a CLI that doesn't exist instead of using the MCP tools. This means:
+1. The MCP server wasn't copied — check `container/agent-runner/src/atomic-chat-mcp-stdio.ts` exists
+2. The MCP server wasn't registered — check `container/agent-runner/src/index.ts` has the `atomic_chat` entry in `mcpServers`
+3. The allowlist wasn't updated — check `container/agent-runner/src/providers/claude.ts` includes `mcp__atomic_chat__*` in `TOOL_ALLOWLIST`
+4. The container wasn't rebuilt — run `./container/build.sh`
+
+### "Failed to connect to Atomic Chat"
+
+1. Verify the host API is reachable: `curl http://127.0.0.1:1337/v1/models`
+2. Confirm the Local API Server is enabled in Atomic Chat's settings
+3. Check Docker can reach the host: `docker run --rm curlimages/curl curl -s http://host.docker.internal:1337/v1/models`
+4. If using a custom host, check `ATOMIC_CHAT_HOST` in `.env`
+
+### `model not found` / 404 on generate
+
+The model ID passed to `atomic_chat_generate` must exactly match one of the IDs returned by `atomic_chat_list_models`. Ask the agent to list models first, then pick one from that list.
+
+### Slow first response
+
+Atomic Chat lazy-loads models into memory on first use. The initial call may take longer while the model warms up. Subsequent calls against the same model are fast.
+
+### Agent doesn't use Atomic Chat tools
+
+The agent may not know about the tools. Try being explicit: "use the atomic_chat_generate tool with llama3.2-3b-instruct to answer: ..."
+
+### Context window or output size issues
+
+Atomic Chat respects each model's native context length. If you hit limits, pass `max_tokens` explicitly when calling `atomic_chat_generate`, or switch to a model with a larger context window in the Atomic Chat UI.

.claude/skills/add-atomic-chat-tool/atomic-chat-mcp-stdio.ts

@@ -0,0 +1,229 @@
+/**
+ * Atomic Chat MCP Server for NanoClaw
+ * Exposes local Atomic Chat models (OpenAI-compatible, /v1) as tools for the container agent.
+ * Uses host.docker.internal to reach the host's Atomic Chat desktop app from Docker.
+ */
+
+import { McpServer } from '@modelcontextprotocol/sdk/server/mcp.js';
+import { StdioServerTransport } from '@modelcontextprotocol/sdk/server/stdio.js';
+import { z } from 'zod';
+
+import fs from 'fs';
+import path from 'path';
+
+const ATOMIC_CHAT_HOST =
+  process.env.ATOMIC_CHAT_HOST || 'http://host.docker.internal:1337';
+const ATOMIC_CHAT_API_KEY = process.env.ATOMIC_CHAT_API_KEY || '';
+const ATOMIC_CHAT_STATUS_FILE = '/workspace/ipc/atomic_chat_status.json';
+
+function log(msg: string): void {
+  console.error(`[ATOMIC] ${msg}`);
+}
+
+function writeStatus(status: string, detail?: string): void {
+  try {
+    const data = { status, detail, timestamp: new Date().toISOString() };
+    const tmpPath = `${ATOMIC_CHAT_STATUS_FILE}.tmp`;
+    fs.mkdirSync(path.dirname(ATOMIC_CHAT_STATUS_FILE), { recursive: true });
+    fs.writeFileSync(tmpPath, JSON.stringify(data));
+    fs.renameSync(tmpPath, ATOMIC_CHAT_STATUS_FILE);
+  } catch {
+    /* best-effort */
+  }
+}
+
+async function atomicFetch(
+  apiPath: string,
+  options?: RequestInit,
+): Promise<Response> {
+  const url = `${ATOMIC_CHAT_HOST}${apiPath}`;
+  const headers: Record<string, string> = {
+    ...((options?.headers as Record<string, string>) || {}),
+  };
+  if (ATOMIC_CHAT_API_KEY) {
+    headers.Authorization = `Bearer ${ATOMIC_CHAT_API_KEY}`;
+  }
+  const finalOptions: RequestInit = { ...options, headers };
+  try {
+    return await fetch(url, finalOptions);
+  } catch (err) {
+    // Fallback to localhost if host.docker.internal fails
+    if (ATOMIC_CHAT_HOST.includes('host.docker.internal')) {
+      const fallbackUrl = url.replace('host.docker.internal', 'localhost');
+      return await fetch(fallbackUrl, finalOptions);
+    }
+    throw err;
+  }
+}
+
+const server = new McpServer({
+  name: 'atomic_chat',
+  version: '1.0.0',
+});
+
+server.tool(
+  'atomic_chat_list_models',
+  'List all models available in the local Atomic Chat desktop app. Use this to see which models are loaded before calling atomic_chat_generate.',
+  {},
+  async () => {
+    log('Listing models...');
+    writeStatus('listing', 'Listing available models');
+    try {
+      const res = await atomicFetch('/v1/models');
+      if (!res.ok) {
+        return {
+          content: [
+            {
+              type: 'text' as const,
+              text: `Atomic Chat API error: ${res.status} ${res.statusText}`,
+            },
+          ],
+          isError: true,
+        };
+      }
+
+      const data = (await res.json()) as {
+        data?: Array<{ id: string; owned_by?: string }>;
+      };
+      const models = data.data || [];
+
+      if (models.length === 0) {
+        return {
+          content: [
+            {
+              type: 'text' as const,
+              text: 'No models available. Open Atomic Chat on the host and download a model from the Hub.',
+            },
+          ],
+        };
+      }
+
+      const list = models
+        .map((m) => `- ${m.id}${m.owned_by ? ` (${m.owned_by})` : ''}`)
+        .join('\n');
+
+      log(`Found ${models.length} models`);
+      return {
+        content: [
+          { type: 'text' as const, text: `Available models:\n${list}` },
+        ],
+      };
+    } catch (err) {
+      return {
+        content: [
+          {
+            type: 'text' as const,
+            text: `Failed to connect to Atomic Chat at ${ATOMIC_CHAT_HOST}: ${err instanceof Error ? err.message : String(err)}`,
+          },
+        ],
+        isError: true,
+      };
+    }
+  },
+);
+
+server.tool(
+  'atomic_chat_generate',
+  'Send a prompt to a local Atomic Chat model and get a response. Good for cheaper/faster tasks like summarization, translation, or general queries. Use atomic_chat_list_models first to see available models.',
+  {
+    model: z
+      .string()
+      .describe(
+        'The model ID as returned by atomic_chat_list_models (e.g. "llama3.2-3b-instruct")',
+      ),
+    prompt: z.string().describe('The prompt to send to the model'),
+    system: z
+      .string()
+      .optional()
+      .describe('Optional system prompt to set model behavior'),
+    temperature: z
+      .number()
+      .optional()
+      .describe('Sampling temperature (0.0–2.0). Defaults to model default.'),
+    max_tokens: z
+      .number()
+      .optional()
+      .describe('Maximum number of tokens to generate in the response.'),
+  },
+  async (args) => {
+    log(`>>> Generating with ${args.model} (${args.prompt.length} chars)...`);
+    writeStatus('generating', `Generating with ${args.model}`);
+    try {
+      const messages: Array<{ role: string; content: string }> = [];
+      if (args.system) {
+        messages.push({ role: 'system', content: args.system });
+      }
+      messages.push({ role: 'user', content: args.prompt });
+
+      const body: Record<string, unknown> = {
+        model: args.model,
+        messages,
+        stream: false,
+      };
+      if (args.temperature !== undefined) body.temperature = args.temperature;
+      if (args.max_tokens !== undefined) body.max_tokens = args.max_tokens;
+
+      const startedAt = Date.now();
+      const res = await atomicFetch('/v1/chat/completions', {
+        method: 'POST',
+        headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify(body),
+      });
+
+      if (!res.ok) {
+        const errorText = await res.text();
+        return {
+          content: [
+            {
+              type: 'text' as const,
+              text: `Atomic Chat error (${res.status}): ${errorText}`,
+            },
+          ],
+          isError: true,
+        };
+      }
+
+      const data = (await res.json()) as {
+        choices?: Array<{ message?: { content?: string } }>;
+        usage?: {
+          prompt_tokens?: number;
+          completion_tokens?: number;
+          total_tokens?: number;
+        };
+      };
+
+      const response = data.choices?.[0]?.message?.content ?? '';
+      const elapsedSec = ((Date.now() - startedAt) / 1000).toFixed(1);
+      const completionTokens = data.usage?.completion_tokens;
+
+      const meta = `\n\n[${args.model} | ${elapsedSec}s${
+        completionTokens !== undefined ? ` | ${completionTokens} tokens` : ''
+      }]`;
+
+      log(
+        `<<< Done: ${args.model} | ${elapsedSec}s | ${
+          completionTokens ?? '?'
+        } tokens | ${response.length} chars`,
+      );
+      writeStatus(
+        'done',
+        `${args.model} | ${elapsedSec}s | ${completionTokens ?? '?'} tokens`,
+      );
+
+      return { content: [{ type: 'text' as const, text: response + meta }] };
+    } catch (err) {
+      return {
+        content: [
+          {
+            type: 'text' as const,
+            text: `Failed to call Atomic Chat: ${err instanceof Error ? err.message : String(err)}`,
+          },
+        ],
+        isError: true,
+      };
+    }
+  },
+);
+
+const transport = new StdioServerTransport();
+await server.connect(transport);

.claude/skills/add-gcal-tool/SKILL.md

@@ -0,0 +1,210 @@
+---
+name: add-gcal-tool
+description: Add Google Calendar as an MCP tool (list calendars, list/search/create events, free/busy queries) using OneCLI-managed OAuth. Multi-calendar and multi-account supported. Mirrors /add-gmail-tool's stub pattern — no raw credentials ever reach the container; OneCLI injects real tokens at request time.
+---
+
+# Add Google Calendar Tool (OneCLI-native)
+
+This skill wires [`@cocal/google-calendar-mcp`](https://github.com/cocal-com/google-calendar-mcp) into selected agent groups. The MCP server reads stub credentials containing the `onecli-managed` placeholder; the OneCLI gateway intercepts outbound calls to `calendar.googleapis.com` / `oauth2.googleapis.com` and swaps the bearer for the real OAuth token from its vault.
+
+**Why this package (and not gongrzhe's):** `@gongrzhe/server-calendar-autoauth-mcp` only supports the `primary` calendar and exposes 5 tools (no `list_calendars`). `@cocal/google-calendar-mcp` explicitly supports multi-calendar and multi-account, and is actively maintained.
+
+Tools exposed (surfaced as `mcp__calendar__<name>`, exact set depends on version — run `tools/list` against the MCP server to enumerate): `list-calendars`, `list-events`, `search-events`, `create-event`, `update-event`, `delete-event`, `get-event`, `list-colors`, `get-freebusy`, `get-current-time`, plus multi-account management tools.
+
+**Why this pattern:** v2's invariant is that containers never receive raw API keys (CHANGELOG 2.0.0). Same stub pattern `/add-gmail-tool` uses. This skill is deliberately a sibling, not a combined "Google Workspace" skill — installs independently and removes cleanly.
+
+## Phase 1: Pre-flight
+
+### Verify OneCLI has Google Calendar connected
+
+```bash
+onecli apps get --provider google-calendar
+```
+
+Expected: `"connection": { "status": "connected" }` with scopes including `calendar.readonly` and `calendar.events`.
+
+If not connected, tell the user:
+
+> Open the OneCLI web UI at http://127.0.0.1:10254, go to Apps → Google Calendar, and click Connect. Sign in with the Google account the agent should act as. `calendar.readonly` + `calendar.events` are the minimum useful scopes.
+
+### Verify stub credentials exist
+
+The stub lives at `~/.calendar-mcp/` by convention (shared with `/add-gmail-tool`'s sibling). cocal doesn't default to this path (it uses `~/.config/google-calendar-mcp/tokens.json`) — we override via env vars below so it reads our stubs instead.
+
+```bash
+ls -la ~/.calendar-mcp/gcp-oauth.keys.json ~/.calendar-mcp/credentials.json 2>&1
+```
+
+If both exist with `onecli-managed`:
+
+```bash
+grep -l onecli-managed ~/.calendar-mcp/gcp-oauth.keys.json ~/.calendar-mcp/credentials.json
+```
+
+...skip to Phase 2. If either file has real credentials (no `onecli-managed`), **STOP** — back up and delete before proceeding.
+
+If absent, write them:
+
+```bash
+mkdir -p ~/.calendar-mcp
+cat > ~/.calendar-mcp/gcp-oauth.keys.json <<'EOF'
+{
+  "installed": {
+    "client_id": "onecli-managed.apps.googleusercontent.com",
+    "client_secret": "onecli-managed",
+    "redirect_uris": ["http://localhost:3000/oauth2callback"]
+  }
+}
+EOF
+cat > ~/.calendar-mcp/credentials.json <<'EOF'
+{
+  "access_token": "onecli-managed",
+  "refresh_token": "onecli-managed",
+  "token_type": "Bearer",
+  "expiry_date": 99999999999999,
+  "scope": "https://www.googleapis.com/auth/calendar.readonly https://www.googleapis.com/auth/calendar.events"
+}
+EOF
+chmod 600 ~/.calendar-mcp/*.json
+```
+
+### Verify mount allowlist covers the path
+
+```bash
+cat ~/.config/nanoclaw/mount-allowlist.json
+```
+
+`~/.calendar-mcp` must sit under an `allowedRoots` entry.
+
+### Check agent secret-mode
+
+For each target agent group, confirm OneCLI will inject the Google Calendar token:
+
+```bash
+onecli agents list
+```
+
+`secretMode: all` is sufficient. If `selective`, explicitly assign the Calendar secret.
+
+## Phase 2: Apply Code Changes
+
+### Check if already applied
+
+```bash
+grep -q 'CALENDAR_MCP_VERSION' container/Dockerfile && \
+grep -q "mcp__calendar__\*" container/agent-runner/src/providers/claude.ts && \
+echo "ALREADY APPLIED — skip to Phase 3"
+```
+
+### Add MCP server to Dockerfile
+
+Edit `container/Dockerfile`. Find the pinned-version ARG block and add:
+
+```dockerfile
+ARG CALENDAR_MCP_VERSION=2.6.1
+```
+
+If `/add-gmail-tool` has already been applied, the pnpm global-install block already exists with its `zod-to-json-schema@3.22.5` pin. Just append the calendar package — **the calendar-mcp uses `zod@4.x` and does NOT need that pin**, but it's harmless to share the block:
+
+```dockerfile
+RUN --mount=type=cache,target=/root/.cache/pnpm \
+    pnpm install -g \
+        "@gongrzhe/server-gmail-autoauth-mcp@${GMAIL_MCP_VERSION}" \
+        "@cocal/google-calendar-mcp@${CALENDAR_MCP_VERSION}" \
+        "zod-to-json-schema@3.22.5"
+```
+
+If `/add-gmail-tool` hasn't been applied, install Calendar standalone:
+
+```dockerfile
+RUN --mount=type=cache,target=/root/.cache/pnpm \
+    pnpm install -g "@cocal/google-calendar-mcp@${CALENDAR_MCP_VERSION}"
+```
+
+### Add tools to allowlist
+
+Edit `container/agent-runner/src/providers/claude.ts`. Add `'mcp__calendar__*'` to `TOOL_ALLOWLIST` after `'mcp__nanoclaw__*'` (or after `'mcp__gmail__*'` if present).
+
+### Rebuild the container image
+
+```bash
+./container/build.sh
+```
+
+## Phase 3: Wire Per-Agent-Group
+
+For each agent group, merge into `groups/<folder>/container.json`:
+
+```jsonc
+{
+  "mcpServers": {
+    "calendar": {
+      "command": "google-calendar-mcp",
+      "args": [],
+      "env": {
+        "GOOGLE_OAUTH_CREDENTIALS": "/workspace/extra/.calendar-mcp/gcp-oauth.keys.json",
+        "GOOGLE_CALENDAR_MCP_TOKEN_PATH": "/workspace/extra/.calendar-mcp/credentials.json"
+      }
+    }
+  },
+  "additionalMounts": [
+    {
+      "hostPath": "/home/<user>/.calendar-mcp",
+      "containerPath": ".calendar-mcp",
+      "readonly": false
+    }
+  ]
+}
+```
+
+Substitute `<user>` with `echo $HOME`. `containerPath` is relative (mount-security rejects absolute paths — additional mounts land at `/workspace/extra/<relative>`).
+
+**Same-group-as-gmail tip:** if this group already has the gmail MCP + `.gmail-mcp` mount, **merge, don't replace** — both entries coexist in `mcpServers` and `additionalMounts`.
+
+## Phase 4: Build and Restart
+
+```bash
+pnpm run build
+systemctl --user restart nanoclaw   # Linux
+# launchctl kickstart -k gui/$(id -u)/com.nanoclaw   # macOS
+```
+
+Kill any existing agent containers so they respawn with the new mcpServers config:
+
+```bash
+docker ps -q --filter 'name=nanoclaw-v2-' | xargs -r docker kill
+```
+
+## Phase 5: Verify
+
+### Test from a wired agent
+
+> Send: **"list my calendars"** or **"what's on my work calendar next Monday?"**.
+>
+> First call takes 2–3s while the MCP server starts and OneCLI does the token exchange.
+
+### Check logs if the tool isn't working
+
+```bash
+tail -100 logs/nanoclaw.log | grep -iE 'calendar|mcp'
+```
+
+Common signals:
+- `command not found: google-calendar-mcp` → image not rebuilt.
+- `ENOENT ...credentials.json` → mount missing. Check the mount allowlist.
+- `401 Unauthorized` from `*.googleapis.com` → OneCLI isn't injecting; verify agent's secret mode and that Google Calendar is connected.
+- Agent says "I don't have calendar tools" → `mcp__calendar__*` missing from `TOOL_ALLOWLIST`, or image cache stale (`./container/build.sh` again).
+
+## Removal
+
+1. Delete `"calendar"` from `mcpServers` and the `.calendar-mcp` mount from `additionalMounts` in each group's `container.json`.
+2. Remove `'mcp__calendar__*'` from `TOOL_ALLOWLIST`.
+3. Remove `CALENDAR_MCP_VERSION` ARG and the calendar package from the Dockerfile install block.
+4. `pnpm run build && ./container/build.sh && systemctl --user restart nanoclaw`.
+5. Optional: `rm -rf ~/.calendar-mcp/` and `onecli apps disconnect --provider google-calendar`.
+
+## Credits & references
+
+- **MCP server:** [`@cocal/google-calendar-mcp`](https://github.com/cocal-com/google-calendar-mcp) — MIT-licensed, actively maintained, multi-account and multi-calendar.
+- **Why not gongrzhe:** earlier versions of this skill used `@gongrzhe/server-calendar-autoauth-mcp@1.0.2` which only supports the primary calendar with 5 event-level tools. The cocal server supersedes it.
+- **Skill pattern:** direct sibling of [`/add-gmail-tool`](../add-gmail-tool/SKILL.md); same OneCLI stub mechanism.

.claude/skills/add-gmail-tool/SKILL.md

@@ -0,0 +1,229 @@
+---
+name: add-gmail-tool
+description: Add Gmail as an MCP tool (read, search, send, label, draft) using OneCLI-managed OAuth. The agent gets Gmail tools in every enabled group; OneCLI injects real tokens at request time so no raw credentials are ever in the container or on disk in usable form.
+---
+
+# Add Gmail Tool (OneCLI-native)
+
+This skill wires the [`@gongrzhe/server-gmail-autoauth-mcp`](https://www.npmjs.com/package/@gongrzhe/server-gmail-autoauth-mcp) stdio MCP server into selected agent groups. The MCP server reads stub credentials containing the `onecli-managed` placeholder; the OneCLI gateway intercepts outbound calls to `gmail.googleapis.com` and injects the real OAuth bearer from its vault.
+
+Tools exposed (from `gmail-mcp@1.1.11`, surfaced to the agent as `mcp__gmail__<name>`): `search_emails`, `read_email`, `send_email`, `draft_email`, `delete_email`, `modify_email`, `batch_modify_emails`, `batch_delete_emails`, `download_attachment`, `list_email_labels`, `create_label`, `update_label`, `delete_label`, `get_or_create_label`, `list_filters`, `get_filter`, `create_filter`, `create_filter_from_template`, `delete_filter`.
+
+**Why this pattern:** v2's invariant is that containers never receive raw API keys — OneCLI is the sole credential path (see CHANGELOG v2.0.0). The stub-file pattern satisfies this: the container sees `"onecli-managed"` placeholders, the gateway swaps them in flight.
+
+## Phase 1: Pre-flight
+
+### Verify OneCLI has Gmail connected
+
+```bash
+onecli apps get --provider gmail
+```
+
+Expected: `"connection": { "status": "connected" }` with scopes including `gmail.readonly`, `gmail.modify`, `gmail.send`.
+
+If not connected, tell the user:
+
+> Open the OneCLI web UI at http://127.0.0.1:10254, go to Apps → Gmail, and click Connect. Sign in with the Google account you want the agent to act as.
+
+### Verify stub credentials exist
+
+```bash
+ls -la ~/.gmail-mcp/gcp-oauth.keys.json ~/.gmail-mcp/credentials.json 2>&1
+```
+
+If both exist and contain `"onecli-managed"`:
+
+```bash
+grep -l onecli-managed ~/.gmail-mcp/gcp-oauth.keys.json ~/.gmail-mcp/credentials.json
+```
+
+...skip to Phase 2.
+
+If either file exists but does **not** contain `onecli-managed`, **STOP** and tell the user — these are real OAuth credentials from a previous non-OneCLI install. Back them up, then delete before proceeding. The OneCLI migration normally handles this; if it didn't, something is wrong.
+
+If both files are absent, write them now:
+
+```bash
+mkdir -p ~/.gmail-mcp
+cat > ~/.gmail-mcp/gcp-oauth.keys.json <<'EOF'
+{
+  "installed": {
+    "client_id": "onecli-managed.apps.googleusercontent.com",
+    "client_secret": "onecli-managed",
+    "redirect_uris": ["http://localhost:3000/oauth2callback"]
+  }
+}
+EOF
+cat > ~/.gmail-mcp/credentials.json <<'EOF'
+{
+  "access_token": "onecli-managed",
+  "refresh_token": "onecli-managed",
+  "token_type": "Bearer",
+  "expiry_date": 99999999999999,
+  "scope": "https://www.googleapis.com/auth/gmail.readonly https://www.googleapis.com/auth/gmail.modify https://www.googleapis.com/auth/gmail.send"
+}
+EOF
+chmod 600 ~/.gmail-mcp/gcp-oauth.keys.json ~/.gmail-mcp/credentials.json
+```
+
+### Verify mount allowlist covers the path
+
+```bash
+cat ~/.config/nanoclaw/mount-allowlist.json
+```
+
+`~/.gmail-mcp` must sit under an `allowedRoots` entry (e.g. `/home/<user>`). If it doesn't, tell the user to run `/manage-mounts` first or add their home directory.
+
+### Check agent secret-mode
+
+For each target agent group, confirm OneCLI will inject Gmail secrets into its container. Find the OneCLI agent ID that matches the group's `agentGroupId`:
+
+```bash
+onecli agents list
+```
+
+If that agent's `secretMode` is `all`, you're done — Gmail secrets (identified by OneCLI's Gmail hostPattern) will auto-inject. If it's `selective`, explicitly assign the Gmail secrets:
+
+```bash
+onecli secrets list     # find Gmail secret IDs (OneCLI creates one per connected app)
+onecli agents set-secrets --id <agent-id> --secret-ids <gmail-secret-id>
+```
+
+## Phase 2: Apply Code Changes
+
+### Check if already applied
+
+```bash
+grep -q 'GMAIL_MCP_VERSION' container/Dockerfile && \
+grep -q "mcp__gmail__\*" container/agent-runner/src/providers/claude.ts && \
+echo "ALREADY APPLIED — skip to Phase 3"
+```
+
+### Add MCP server to Dockerfile
+
+Edit `container/Dockerfile`. Find the pinned-version ARG block:
+
+```dockerfile
+ARG CLAUDE_CODE_VERSION=2.1.116
+ARG AGENT_BROWSER_VERSION=latest
+ARG VERCEL_VERSION=latest
+ARG BUN_VERSION=1.3.12
+```
+
+Add a new line:
+
+```dockerfile
+ARG GMAIL_MCP_VERSION=1.1.11
+```
+
+Then find the last pnpm global-install `RUN` block (the one that installs `@anthropic-ai/claude-code`) and add a new block after it, before `# ---- Entrypoint`:
+
+```dockerfile
+RUN --mount=type=cache,target=/root/.cache/pnpm \
+    pnpm install -g \
+        "@gongrzhe/server-gmail-autoauth-mcp@${GMAIL_MCP_VERSION}" \
+        "zod-to-json-schema@3.22.5"
+```
+
+Pinned version matters — `minimumReleaseAge` in `pnpm-workspace.yaml` gates trunk installs, and CLAUDE.md requires a fixed ARG version for all Node CLIs installed into the image.
+
+**Why the `zod-to-json-schema` pin:** `@gongrzhe/server-gmail-autoauth-mcp@1.1.11` has loose deps (`zod-to-json-schema: ^3.22.1`, `zod: ^3.22.4`). pnpm resolves `zod-to-json-schema` to the latest 3.25.x, which imports `zod/v3` — a subpath that only exists in `zod>=3.25`. But `zod` resolves to `3.24.x` (highest satisfying `^3.22.4` without breaking peer ranges). Result: `ERR_PACKAGE_PATH_NOT_EXPORTED` at import time. Pinning `zod-to-json-schema` to a pre-v3-subpath version avoids it. Re-check if you bump `GMAIL_MCP_VERSION`.
+
+### Add tools to allowlist
+
+Edit `container/agent-runner/src/providers/claude.ts`. Find `'mcp__nanoclaw__*',` in `TOOL_ALLOWLIST` and add `'mcp__gmail__*',` after it.
+
+### Rebuild the container image
+
+```bash
+./container/build.sh
+```
+
+Must complete cleanly. The new `pnpm install -g` layer is ~60s first time (cached on rebuild).
+
+## Phase 3: Wire Per-Agent-Group
+
+For each agent group that should have Gmail (ask the user — typically their personal DM and CLI agents, sometimes shared household agents), edit `groups/<folder>/container.json` to add the mount and MCP server.
+
+Merge these into the group's `container.json`:
+
+```jsonc
+{
+  "mcpServers": {
+    "gmail": {
+      "command": "gmail-mcp",
+      "args": [],
+      "env": {
+        "GMAIL_OAUTH_PATH": "/workspace/extra/.gmail-mcp/gcp-oauth.keys.json",
+        "GMAIL_CREDENTIALS_PATH": "/workspace/extra/.gmail-mcp/credentials.json"
+      }
+    }
+  },
+  "additionalMounts": [
+    {
+      "hostPath": "/home/<user>/.gmail-mcp",
+      "containerPath": ".gmail-mcp",
+      "readonly": false
+    }
+  ]
+}
+```
+
+Substitute `<user>` with the host user's home (use `echo $HOME`, don't assume `~` will expand — `container-runner.ts` does expand `~` via `expandPath`, but an explicit absolute path is clearer and matches what `/manage-mounts` writes).
+
+**Why the container path is relative:** `mount-security` rejects absolute `containerPath` values. Additional mounts are prefixed with `/workspace/extra/`, so `containerPath: ".gmail-mcp"` lands at `/workspace/extra/.gmail-mcp`. The MCP server's `GMAIL_OAUTH_PATH` / `GMAIL_CREDENTIALS_PATH` env vars point at that absolute location inside the container.
+
+## Phase 4: Build and Restart
+
+```bash
+pnpm run build
+systemctl --user restart nanoclaw  # Linux
+# launchctl kickstart -k gui/$(id -u)/com.nanoclaw   # macOS
+```
+
+## Phase 5: Verify
+
+### Test from the wired agent
+
+Tell the user:
+
+> In your `<agent-name>` chat, send: **"list my gmail labels"** or **"search my inbox for invoices from last month"**.
+>
+> The agent should use `mcp__gmail__list_labels` / `mcp__gmail__search`. The first call may take a second or two while the MCP server starts and OneCLI does the token exchange.
+
+### Check logs if the tool isn't working
+
+```bash
+tail -100 logs/nanoclaw.log logs/nanoclaw.error.log | grep -iE 'gmail|mcp'
+# Per-container logs — session-scoped:
+ls data/v2-sessions/*/stderr.log | head
+```
+
+Common signals:
+- `command not found: gmail-mcp` → image wasn't rebuilt or PATH doesn't include `/pnpm` (should — `ENV PATH="$PNPM_HOME:$PATH"` in Dockerfile).
+- `ENOENT: no such file or directory, open '/workspace/extra/.gmail-mcp/credentials.json'` → mount is missing. Check `~/.config/nanoclaw/mount-allowlist.json` includes a parent of `~/.gmail-mcp`.
+- `401 Unauthorized` from `gmail.googleapis.com` → OneCLI isn't injecting. Check the agent's secret mode (`onecli agents secrets --id <agent-id>`) and that the Gmail app is connected (`onecli apps get --provider gmail`).
+- Agent says "I don't have Gmail tools" → `mcp__gmail__*` wasn't added to `TOOL_ALLOWLIST`, or the agent-runner wasn't rebuilt (image cache — run `./container/build.sh` again with `--no-cache` if suspicious).
+
+## Removal
+
+1. Delete the `"gmail"` entry from `mcpServers` and the `.gmail-mcp` entry from `additionalMounts` in each group's `container.json`.
+2. Remove `'mcp__gmail__*'` from `TOOL_ALLOWLIST` in `container/agent-runner/src/providers/claude.ts`.
+3. Remove the `GMAIL_MCP_VERSION` ARG and the `pnpm install -g @gongrzhe/server-gmail-autoauth-mcp` block from `container/Dockerfile`.
+4. `pnpm run build && ./container/build.sh && systemctl --user restart nanoclaw`.
+5. (Optional) `rm -rf ~/.gmail-mcp/` if no other host-side tool needs the stubs.
+6. (Optional) Disconnect Gmail in OneCLI: `onecli apps disconnect --provider gmail`.
+
+## Notes
+
+- **Stub format is OneCLI-prescribed.** The `access_token: "onecli-managed"` pattern with `expiry_date: 99999999999999` tells the Google auth client the token is valid; OneCLI intercepts the outgoing Gmail API call and rewrites `Authorization: Bearer onecli-managed` to the real token. `expiry_date: 0` (refresh-interception) is an alternative the OneCLI docs describe — both work but OneCLI's own `migrate` command writes the far-future variant, which is what this skill assumes.
+- **Scopes are set at OAuth connect time.** If the agent needs scopes beyond what's currently connected (e.g. the user later wants `calendar.readonly` for combined email/calendar workflows), disconnect and reconnect Gmail in the OneCLI web UI with the expanded scope set.
+- **This is tool-only.** Inbound email as a channel (emails trigger the agent) is a separate piece of work — it needs a `src/channels/gmail.ts` adapter that polls the inbox and routes to a messaging group. The pre-v2 qwibitai skill had this; it has not been ported to v2's channel architecture as of v2.0.0.
+
+## Credits & references
+
+- **MCP server:** [`@gongrzhe/server-gmail-autoauth-mcp`](https://github.com/GongRzhe/Gmail-MCP-Server) by GongRzhe — MIT-licensed.
+- **OneCLI credential stubs:** pattern documented at `https://onecli.sh/docs/guides/credential-stubs/gmail.md`.
+- **Skill pattern:** modeled on [`add-atomic-chat-tool`](../add-atomic-chat-tool/SKILL.md) and [`add-vercel`](../add-vercel/SKILL.md).
+- **Addresses:** [issue #1500](https://github.com/qwibitai/nanoclaw/issues/1500) (proxy Gmail/Calendar OAuth tokens through credential proxy) for the Gmail side.
+- **Related PRs:** [#1810](https://github.com/qwibitai/nanoclaw/pull/1810) (pre-install Gmail/Notion MCP) overlaps on the "install the MCP server in the image" idea but bundles many unrelated changes; this skill is the focused OneCLI-native version.

.claude/skills/add-opencode/SKILL.md

@@ -208,7 +208,7 @@ onecli secrets create --name "OpenCode Zen" --type generic \
 
 ### Per group / per session
 
-Schema: **`agent_groups.agent_provider`** and **`sessions.agent_provider`**. Set to `opencode` for groups or sessions that should use OpenCode. The container receives `AGENT_PROVIDER` from the resolved value (session overrides group).
+Set `"provider": "opencode"` in the group's **`container.json`** (`groups/<folder>/container.json`) — the in-container runner reads `provider` from there, not from the DB. The DB columns **`agent_groups.agent_provider`** and **`sessions.agent_provider`** (session overrides group) only drive host-side provider contribution — per-session XDG mount, `OPENCODE_*` env passthrough — and do not propagate into `container.json` at spawn time. Set both, or just edit `container.json`; if they disagree, the runner uses `container.json` and the host-side resolver falls back through session → group → `container.json` → `'claude'`.
 
 Extra MCP servers still come from **`NANOCLAW_MCP_SERVERS`** / `container_config.mcpServers` on the host; the runner merges them into the same `mcpServers` object passed to **both** Claude and OpenCode providers.
 

.claude/skills/add-signal/REMOVE.md

@@ -0,0 +1,13 @@
+# Remove Signal
+
+1. Comment out `import './signal.js'` in `src/channels/index.ts`
+2. Remove `SIGNAL_ACCOUNT` (and any other `SIGNAL_*` vars) from `.env`
+3. Rebuild and restart
+
+If you also want to unlink the Signal account from `signal-cli`:
+
+```bash
+signal-cli -a +1YOURNUMBER removeDevice --deviceId <id>
+```
+
+(Find the device id with `signal-cli -a +1YOURNUMBER listDevices`.)

.claude/skills/add-signal/SKILL.md

@@ -0,0 +1,318 @@
+---
+name: add-signal
+description: Add Signal channel integration via signal-cli TCP daemon. Native adapter — no Chat SDK bridge.
+---
+
+# Add Signal Channel
+
+Adds Signal messaging support via a native adapter that speaks JSON-RPC to a [signal-cli](https://github.com/AsamK/signal-cli) TCP daemon. No Chat SDK bridge — only Node.js builtins (`node:net`, `node:child_process`, `node:fs`).
+
+Unlike Telegram or Discord, Signal has no bot API. NanoClaw registers as a full Signal account on a dedicated phone number (recommended) or links as a secondary device on your existing number.
+
+## Prerequisites
+
+### Java
+
+signal-cli requires Java 17+:
+
+```bash
+java -version
+```
+
+If missing:
+- **macOS:** `brew install --cask temurin@17`
+- **Debian/Ubuntu:** `sudo apt-get install -y default-jre`
+- **RHEL/Fedora:** `sudo dnf install -y java-17-openjdk`
+
+Java 17–25 all work.
+
+### signal-cli
+
+- **macOS:** `brew install signal-cli`
+- **Linux:** download the native binary from [GitHub releases](https://github.com/AsamK/signal-cli/releases):
+
+```bash
+SIGNAL_CLI_VERSION=$(curl -fsSL https://api.github.com/repos/AsamK/signal-cli/releases/latest | python3 -c "import sys,json; print(json.load(sys.stdin)['tag_name'][1:])")
+curl -fsSL "https://github.com/AsamK/signal-cli/releases/download/v${SIGNAL_CLI_VERSION}/signal-cli-${SIGNAL_CLI_VERSION}-Linux-native.tar.gz" \
+  | tar -xz -C ~/.local
+ln -sf ~/.local/signal-cli ~/.local/bin/signal-cli
+signal-cli --version
+```
+
+> The Linux native tarball extracts a single binary directly to `~/.local/signal-cli` (not into a subdirectory). The symlink above puts it on PATH.
+
+## Registration
+
+Two paths. The new-number path is recommended and battle-tested.
+
+### Path A: Register a new number (recommended)
+
+Use a dedicated SIM or VoIP number. NanoClaw owns it entirely.
+
+> **VoIP numbers:** Signal requires SMS verification before voice. Some VoIP providers are blocked even for voice calls. If registration fails with an auth error, try a different provider or a physical SIM.
+
+**Step 1: Solve the CAPTCHA**
+
+Signal requires a CAPTCHA on first registration:
+
+1. Open `https://signalcaptchas.org/registration/generate.html` in a browser
+2. Solve the captcha
+3. Right-click the **"Open Signal"** button → **Copy Link**
+4. The link starts with `signalcaptcha://` — the token is everything after that prefix
+
+**Step 2: Request SMS verification**
+
+```bash
+signal-cli -a +1YOURNUMBER register --captcha "PASTE_TOKEN_HERE"
+```
+
+**Step 3: Voice call fallback (if your number can't receive SMS)**
+
+Wait ~60 seconds after the SMS request, then:
+
+```bash
+signal-cli -a +1YOURNUMBER register --voice --captcha "SAME_TOKEN"
+```
+
+Signal calls your number and reads a 6-digit code. The same captcha token is reusable — no need to solve a new one.
+
+> You must request SMS first. Requesting voice immediately fails with `Invalid verification method: Before requesting voice verification…`
+
+**Step 4: Verify**
+
+```bash
+signal-cli -a +1YOURNUMBER verify CODE
+```
+
+No output = success.
+
+**Step 5: Set profile name (optional)**
+
+> ⚠ Stop NanoClaw before running signal-cli commands — the daemon holds an exclusive lock on its data directory while running.
+
+```bash
+# macOS
+launchctl unload ~/Library/LaunchAgents/com.nanoclaw.plist
+signal-cli -a +1YOURNUMBER updateProfile --name "YourBotName"
+# optionally: --avatar /path/to/avatar.jpg
+launchctl load ~/Library/LaunchAgents/com.nanoclaw.plist
+
+# Linux
+systemctl --user stop nanoclaw
+signal-cli -a +1YOURNUMBER updateProfile --name "YourBotName"
+systemctl --user start nanoclaw
+```
+
+### Path B: Link as secondary device
+
+Joins an existing Signal account as a secondary device. Simpler, but NanoClaw shares your personal number.
+
+```bash
+signal-cli -a +1YOURNUMBER link --name "NanoClaw"
+```
+
+This prints a `tsdevice:` URI. Scan it as a QR code on your phone: **Settings → Linked Devices → Link New Device**. QR codes expire in ~30 seconds — re-run if it expires.
+
+## Install
+
+### Pre-flight (idempotent)
+
+Skip to **Credentials** if all of these are already in place:
+
+- `src/channels/signal.ts` and `src/channels/signal.test.ts` both exist
+- `src/channels/index.ts` contains `import './signal.js';`
+
+Otherwise continue. Every step below is safe to re-run.
+
+### 1. Fetch the channels branch
+
+```bash
+git fetch origin channels
+```
+
+### 2. Copy the adapter and tests
+
+```bash
+git show origin/channels:src/channels/signal.ts      > src/channels/signal.ts
+git show origin/channels:src/channels/signal.test.ts > src/channels/signal.test.ts
+```
+
+### 3. Append the self-registration import
+
+Append to `src/channels/index.ts` (skip if the line is already present):
+
+```typescript
+import './signal.js';
+```
+
+### 4. Build
+
+```bash
+pnpm run build
+```
+
+No npm packages to install — the adapter uses only Node.js builtins.
+
+## Credentials
+
+Add to `.env`:
+
+```bash
+SIGNAL_ACCOUNT=+1YOURNUMBER
+```
+
+### Optional settings
+
+```bash
+# TCP daemon host and port (default: 127.0.0.1:7583)
+SIGNAL_TCP_HOST=127.0.0.1
+SIGNAL_TCP_PORT=7583
+
+# Path to the signal-cli binary (default: resolved on PATH)
+SIGNAL_CLI_PATH=/usr/local/bin/signal-cli
+
+# Whether NanoClaw manages the daemon lifecycle (default: true).
+# Set to false if you run signal-cli daemon externally.
+SIGNAL_MANAGE_DAEMON=true
+
+# signal-cli data directory (default: ~/.local/share/signal-cli)
+SIGNAL_DATA_DIR=~/.local/share/signal-cli
+```
+
+**Security note:** keep the TCP host on `127.0.0.1`. The daemon has no auth — binding it to a public interface would expose your full Signal account to the network.
+
+Sync to container: `mkdir -p data/env && cp .env data/env/env`
+
+### Restart
+
+```bash
+# macOS
+launchctl kickstart -k gui/$(id -u)/com.nanoclaw
+
+# Linux
+systemctl --user restart nanoclaw
+```
+
+## Wiring
+
+### DMs
+
+After the service starts, send any message to the Signal number from your personal Signal app. The router auto-creates a `messaging_groups` row. Then:
+
+```bash
+sqlite3 data/v2.db \
+  "SELECT id, platform_id FROM messaging_groups WHERE channel_type='signal' ORDER BY created_at DESC LIMIT 5"
+```
+
+Pass the `id` to `/init-first-agent` or `/manage-channels` to wire it to an agent group.
+
+### Groups
+
+Add the Signal number to a group from your phone, send any message, then wire the resulting row the same way. For isolated per-group sessions:
+
+```bash
+NOW=$(date -u +"%Y-%m-%dT%H:%M:%S.000Z")
+sqlite3 data/v2.db "
+INSERT OR IGNORE INTO messaging_group_agents
+  (id, messaging_group_id, agent_group_id, session_mode, priority, created_at)
+VALUES
+  ('mga-'||hex(randomblob(8)), 'mg-GROUPID', 'ag-AGENTID', 'isolated', 0, '$NOW');
+"
+```
+
+### Grant user access
+
+New Signal users (including the owner's Signal identity) are silently dropped with `not_member` until granted access. After the user's first message appears in `messaging_groups`:
+
+```bash
+NOW=$(date -u +"%Y-%m-%dT%H:%M:%S.000Z")
+sqlite3 data/v2.db "
+INSERT OR REPLACE INTO user_roles (user_id, role, agent_group_id, granted_by, granted_at)
+  VALUES ('signal:UUID', 'owner', NULL, 'system', '$NOW');
+INSERT OR IGNORE INTO agent_group_members (user_id, agent_group_id, added_by, added_at)
+  VALUES ('signal:UUID', 'ag-AGENTID', 'system', '$NOW');
+"
+```
+
+Find the UUID from `messaging_groups.platform_id` or the `users` table.
+
+## Next Steps
+
+If you're in the middle of `/setup`, return to the setup flow now.
+
+Otherwise, run `/init-first-agent` to create an agent and wire it to your Signal DM, or `/manage-channels` to wire this channel to an existing agent group.
+
+## Channel Info
+
+- **type**: `signal`
+- **terminology**: Signal has "chats" (1:1 DMs) and "groups"
+- **supports-threads**: no
+- **platform-id-format**:
+  - DM: `signal:{UUID}` — sender's Signal UUID (ACI), **not** their phone number
+  - Group: `signal:{base64GroupId}` — base64-encoded GroupV2 ID
+- **how-to-find-id**: Send a message to the bot, then query `messaging_groups` as shown above
+- **typical-use**: Personal assistant via Signal DMs or small group chats
+- **default-isolation**: One agent per Signal account. Multiple chats with the same operator can share an agent group; groups with other people should typically use `isolated` session mode
+
+### Features
+
+- Markdown formatting — `**bold**`, `*italic*` / `_italic_`, `` `code` ``, ` ```code fence``` `, `~~strike~~`, `||spoiler||` (converted to Signal's offset-based text styles)
+- Quoted replies — `replyTo*` fields populated from Signal quotes
+- Typing indicators — DMs only (Signal doesn't support group typing)
+- Echo suppression — outbound messages matched on `(platformId, text)` within a 10 s TTL to avoid syncMessage loops
+- Note to Self — messages you send to your own account from another device route to the agent as inbound with `isFromMe: true`
+- Voice attachments — detected but not transcribed by default; the agent receives `[Voice Message]` placeholder text. Run `/add-voice-transcription` for local transcription via parakeet-mlx
+
+Not supported yet: outbound file attachments (logged and dropped), edit/delete messages, reactions.
+
+## Troubleshooting
+
+### Daemon not reachable
+
+```bash
+grep "Signal" logs/nanoclaw.log | tail
+```
+
+If you see `Signal daemon failed to start. Is signal-cli installed and your account linked?`:
+- Confirm `signal-cli` is on PATH (or set `SIGNAL_CLI_PATH`)
+- Confirm the account is linked: `signal-cli -a +1YOURNUMBER listIdentities` should succeed without prompting
+
+If you see `Signal daemon not reachable at 127.0.0.1:7583` and `SIGNAL_MANAGE_DAEMON=false`, start the daemon yourself: `signal-cli -a +1YOURNUMBER daemon --tcp 127.0.0.1:7583`.
+
+### Bot not responding
+
+1. Channel initialized: `grep "Signal channel connected" logs/nanoclaw.log | tail -1`
+2. Channel wired: `sqlite3 data/v2.db "SELECT mg.platform_id, mg.name FROM messaging_groups mg JOIN messaging_group_agents mga ON mg.id = mga.messaging_group_id WHERE mg.channel_type='signal'"`
+3. Service running: `launchctl print gui/$(id -u)/com.nanoclaw` (macOS) / `systemctl --user status nanoclaw` (Linux)
+
+### Lost connection mid-session
+
+If you see `Signal channel lost TCP connection to signal-cli daemon` in the logs, the daemon dropped the connection. Restart the service to re-establish.
+
+### Messages dropped with `not_member`
+
+The Signal user hasn't been granted membership. See "Grant user access" above. This affects every new Signal user, including the owner's Signal identity — which is a separate user record from their identity on other channels even if it's the same person.
+
+### Captcha required
+
+Signal requires a captcha for new registrations. Go to `https://signalcaptchas.org/registration/generate.html`, solve it, right-click "Open Signal", copy the link, extract the token after `signalcaptcha://`.
+
+### `Invalid verification method: Before requesting voice verification…`
+
+You must request SMS first, wait ~60 seconds, then request voice. Both steps can use the same captcha token.
+
+### Config file in use / daemon lock
+
+signal-cli holds an exclusive lock on its data directory while the daemon is running. Stop NanoClaw before running any `signal-cli` commands directly, then restart afterward.
+
+### Group replies going to DM instead of group
+
+Modern Signal groups use GroupV2. The adapter must extract the group ID from `envelope?.dataMessage?.groupV2?.id` — not `groupInfo?.groupId`, which is GroupV1/legacy. If group messages are routing as DMs, check `src/channels/signal.ts` and confirm the groupId extraction falls through to `groupV2.id`.
+
+### Java not found
+
+Install Java 17+ — see the Prerequisites section above.
+
+### QR code expired (Path B)
+
+QR codes expire in ~30 seconds. Re-run the link command to generate a new one.

.claude/skills/add-signal/VERIFY.md

@@ -0,0 +1,5 @@
+# Verify Signal
+
+Send a message to your own Signal number (Note to Self) from another device, or have someone send your linked number a DM. The bot should respond within a few seconds.
+
+If nothing happens, tail `logs/nanoclaw.log` for `Signal channel connected` and `Signal message received`.

.github/workflows/label-pr.yml

@@ -1,7 +1,12 @@
 name: Label PR
 
+# SECURITY: this workflow runs with write access to the base repo on fork PRs,
+# because `pull_request_target` executes in the context of the base branch.
+# Keep it metadata-only — do NOT add actions/checkout or any step that
+# executes PR-supplied content (install scripts, build commands, etc.).
+# See https://securitylab.github.com/resources/github-actions-preventing-pwn-requests/
 on:
-  pull_request:
+  pull_request_target:
     types: [opened, edited]
 
 jobs:

container/AGENTS.md

@@ -0,0 +1,15 @@
+You are a NanoClaw agent. Your name, destinations, and message-sending rules are provided in the runtime system prompt at the top of each turn.
+
+## Communication
+
+Be concise. Prefer outcomes over play-by-play; when the work is done, the final message should be about the result.
+
+When you produce a file for the user in the workspace — a document, export, or asset — deliver it with `send_file` in the same turn; announcing without sending is an unfinished reply.
+
+## Workspace
+
+Files you create are saved in `/workspace/agent/`. Use this for notes, research, artifacts, and anything that should persist across turns in this group.
+
+## Conversation History
+
+The `conversations/` folder holds searchable past conversation transcripts or exchange archives for this group. Use it to recall prior context when a request references something that happened before.

container/agent-runner/bun.lock

@@ -7,6 +7,7 @@
       "dependencies": {
         "@anthropic-ai/claude-agent-sdk": "^0.2.116",
         "@modelcontextprotocol/sdk": "^1.12.1",
+        "@opencode-ai/sdk": "^1.4.3",
         "cron-parser": "^5.0.0",
         "zod": "^4.0.0",
       },
@@ -44,6 +45,8 @@
 
     "@modelcontextprotocol/sdk": ["@modelcontextprotocol/sdk@1.29.0", "", { "dependencies": { "@hono/node-server": "^1.19.9", "ajv": "^8.17.1", "ajv-formats": "^3.0.1", "content-type": "^1.0.5", "cors": "^2.8.5", "cross-spawn": "^7.0.5", "eventsource": "^3.0.2", "eventsource-parser": "^3.0.0", "express": "^5.2.1", "express-rate-limit": "^8.2.1", "hono": "^4.11.4", "jose": "^6.1.3", "json-schema-typed": "^8.0.2", "pkce-challenge": "^5.0.0", "raw-body": "^3.0.0", "zod": "^3.25 || ^4.0", "zod-to-json-schema": "^3.25.1" }, "peerDependencies": { "@cfworker/json-schema": "^4.1.1" }, "optionalPeers": ["@cfworker/json-schema"] }, "sha512-zo37mZA9hJWpULgkRpowewez1y6ML5GsXJPY8FI0tBBCd77HEvza4jDqRKOXgHNn867PVGCyTdzqpz0izu5ZjQ=="],
 
+    "@opencode-ai/sdk": ["@opencode-ai/sdk@1.4.11", "", { "dependencies": { "cross-spawn": "7.0.6" } }, "sha512-EJxSfc7D/dda/vrw8zQe4g7yVTxERktvb5SvIBlGBnKYQJGOgo9RyA/1EL3l208rHeo6jm1sdrAF0E6o/k94ug=="],
+
     "@types/bun": ["@types/bun@1.3.12", "", { "dependencies": { "bun-types": "1.3.12" } }, "sha512-DBv81elK+/VSwXHDlnH3Qduw+KxkTIWi7TXkAeh24zpi5l0B2kUg9Ga3tb4nJaPcOFswflgi/yAvMVBPrxMB+A=="],
 
     "@types/node": ["@types/node@22.19.17", "", { "dependencies": { "undici-types": "~6.21.0" } }, "sha512-wGdMcf+vPYM6jikpS/qhg6WiqSV/OhG+jeeHT/KlVqxYfD40iYJf9/AE1uQxVWFvU7MipKRkRv8NSHiCGgPr8Q=="],

container/agent-runner/package.json

@@ -11,6 +11,7 @@
   "dependencies": {
     "@anthropic-ai/claude-agent-sdk": "^0.2.116",
     "@modelcontextprotocol/sdk": "^1.12.1",
+    "@opencode-ai/sdk": "^1.4.3",
     "cron-parser": "^5.0.0",
     "zod": "^4.0.0"
   },

container/agent-runner/src/db/session-state.test.ts

@@ -0,0 +1,100 @@
+import { beforeEach, describe, expect, test } from 'bun:test';
+
+import { getOutboundDb, initTestSessionDb } from './connection.js';
+import {
+  clearContinuation,
+  getContinuation,
+  migrateLegacyContinuation,
+  setContinuation,
+} from './session-state.js';
+
+beforeEach(() => {
+  initTestSessionDb();
+});
+
+function seedLegacy(value: string): void {
+  getOutboundDb()
+    .prepare('INSERT INTO session_state (key, value, updated_at) VALUES (?, ?, ?)')
+    .run('sdk_session_id', value, new Date().toISOString());
+}
+
+describe('session-state — per-provider continuations', () => {
+  test('set/get round-trip, case-insensitive provider key', () => {
+    setContinuation('claude', 'claude-conv-1');
+    expect(getContinuation('claude')).toBe('claude-conv-1');
+    expect(getContinuation('Claude')).toBe('claude-conv-1');
+    expect(getContinuation('CLAUDE')).toBe('claude-conv-1');
+  });
+
+  test('providers are isolated — switching reads the right slot', () => {
+    setContinuation('claude', 'claude-conv-1');
+    setContinuation('codex', 'codex-thread-xyz');
+
+    expect(getContinuation('claude')).toBe('claude-conv-1');
+    expect(getContinuation('codex')).toBe('codex-thread-xyz');
+  });
+
+  test('clearContinuation only affects the specified provider', () => {
+    setContinuation('claude', 'keep-me');
+    setContinuation('codex', 'drop-me');
+
+    clearContinuation('codex');
+
+    expect(getContinuation('claude')).toBe('keep-me');
+    expect(getContinuation('codex')).toBeUndefined();
+  });
+
+  test('unknown provider returns undefined', () => {
+    expect(getContinuation('never-used')).toBeUndefined();
+  });
+});
+
+describe('session-state — legacy migration', () => {
+  test('adopts legacy value into current provider when current is empty', () => {
+    seedLegacy('old-session-id');
+
+    const adopted = migrateLegacyContinuation('claude');
+
+    expect(adopted).toBe('old-session-id');
+    expect(getContinuation('claude')).toBe('old-session-id');
+  });
+
+  test('always deletes legacy row regardless of migration outcome', () => {
+    seedLegacy('old-session-id');
+    setContinuation('claude', 'existing');
+
+    migrateLegacyContinuation('claude');
+
+    // After migration the legacy key must be gone, whether or not it was adopted.
+    // A subsequent migration for a different provider must not see it.
+    const resultAfterSecondCall = migrateLegacyContinuation('codex');
+    expect(resultAfterSecondCall).toBeUndefined();
+  });
+
+  test('prefers existing current-provider slot over legacy', () => {
+    seedLegacy('legacy-value');
+    setContinuation('claude', 'claude-value');
+
+    const result = migrateLegacyContinuation('claude');
+
+    expect(result).toBe('claude-value');
+    expect(getContinuation('claude')).toBe('claude-value');
+  });
+
+  test('no legacy row — returns current provider value (possibly undefined)', () => {
+    expect(migrateLegacyContinuation('claude')).toBeUndefined();
+
+    setContinuation('codex', 'codex-value');
+    expect(migrateLegacyContinuation('codex')).toBe('codex-value');
+  });
+
+  test('migration is idempotent on a second call (legacy already gone)', () => {
+    seedLegacy('once');
+
+    const first = migrateLegacyContinuation('claude');
+    expect(first).toBe('once');
+
+    const second = migrateLegacyContinuation('claude');
+    expect(second).toBe('once');
+  });
+});

container/agent-runner/src/db/session-state.ts

@@ -2,12 +2,20 @@
  * Persistent key/value state for the container. Lives in outbound.db
  * (container-owned, already scoped per channel/thread).
  *
- * Primary use: remember the SDK session ID so the agent's conversation
- * resumes across container restarts. Cleared by /clear.
+ * Primary use: remember each provider's opaque continuation id so the
+ * agent's conversation resumes across container restarts. Keyed per
+ * provider because continuations are provider-private — a Claude
+ * conversation id means nothing to Codex and vice versa. Switching
+ * providers is therefore lossless: each provider's last thread stays
+ * on file and resumes cleanly if the user flips back.
  */
 import { getOutboundDb } from './connection.js';
 
-const SDK_SESSION_KEY = 'sdk_session_id';
+const LEGACY_KEY = 'sdk_session_id';
+
+function continuationKey(providerName: string): string {
+  return `continuation:${providerName.toLowerCase()}`;
+}
 
 function getValue(key: string): string | undefined {
   const row = getOutboundDb()
@@ -18,9 +26,7 @@ function getValue(key: string): string | undefined {
 
 function setValue(key: string, value: string): void {
   getOutboundDb()
-    .prepare(
-      'INSERT OR REPLACE INTO session_state (key, value, updated_at) VALUES (?, ?, ?)',
-    )
+    .prepare('INSERT OR REPLACE INTO session_state (key, value, updated_at) VALUES (?, ?, ?)')
     .run(key, value, new Date().toISOString());
 }
 
@@ -28,14 +34,46 @@ function deleteValue(key: string): void {
   getOutboundDb().prepare('DELETE FROM session_state WHERE key = ?').run(key);
 }
 
-export function getStoredSessionId(): string | undefined {
-  return getValue(SDK_SESSION_KEY);
+/**
+ * One-time migration of the pre-per-provider continuation row.
+ *
+ * Before this was keyed per provider, continuations lived under the
+ * single key `sdk_session_id`. On container start, if that legacy row
+ * exists and the current provider has no continuation of its own, adopt
+ * the legacy value into the current provider's slot (best-guess — the
+ * legacy row was written by whatever provider ran last). The legacy row
+ * is always deleted so future provider flips never re-read a stale id
+ * through the wrong lens.
+ *
+ * Returns the continuation the caller should use at startup (either the
+ * current provider's existing value, the adopted legacy value, or
+ * undefined).
+ */
+export function migrateLegacyContinuation(providerName: string): string | undefined {
+  const legacy = getValue(LEGACY_KEY);
+  const currentKey = continuationKey(providerName);
+  const current = getValue(currentKey);
+
+  if (legacy === undefined) return current;
+
+  // Always drop the legacy row so no future provider reads it.
+  deleteValue(LEGACY_KEY);
+
+  // Prefer the current provider's own slot if one already exists.
+  if (current !== undefined) return current;
+
+  setValue(currentKey, legacy);
+  return legacy;
 }
 
-export function setStoredSessionId(sessionId: string): void {
-  setValue(SDK_SESSION_KEY, sessionId);
+export function getContinuation(providerName: string): string | undefined {
+  return getValue(continuationKey(providerName));
 }
 
-export function clearStoredSessionId(): void {
-  deleteValue(SDK_SESSION_KEY);
+export function setContinuation(providerName: string, id: string): void {
+  setValue(continuationKey(providerName), id);
+}
+
+export function clearContinuation(providerName: string): void {
+  deleteValue(continuationKey(providerName));
 }

container/agent-runner/src/index.ts

@@ -95,6 +95,7 @@ async function main(): Promise<void> {
 
   await runPollLoop({
     provider,
+    providerName,
     cwd: CWD,
     systemContext: { instructions },
   });

container/agent-runner/src/integration.test.ts

@@ -98,6 +98,7 @@ async function runPollLoopWithTimeout(provider: MockProvider, signal: AbortSigna
   return Promise.race([
     runPollLoop({
       provider,
+      providerName: 'mock',
       cwd: '/tmp',
     }),
     new Promise<void>((_, reject) => {

container/agent-runner/src/poll-loop.ts

@@ -2,7 +2,11 @@ import { findByName, getAllDestinations, type DestinationEntry } from './destina
 import { getPendingMessages, markProcessing, markCompleted, type MessageInRow } from './db/messages-in.js';
 import { writeMessageOut } from './db/messages-out.js';
 import { touchHeartbeat, clearStaleProcessingAcks } from './db/connection.js';
-import { getStoredSessionId, setStoredSessionId, clearStoredSessionId } from './db/session-state.js';
+import {
+  clearContinuation,
+  migrateLegacyContinuation,
+  setContinuation,
+} from './db/session-state.js';
 import { formatMessages, extractRouting, categorizeMessage, isClearCommand, stripInternalTags, type RoutingContext } from './formatter.js';
 import type { AgentProvider, AgentQuery, ProviderEvent } from './providers/types.js';
 
@@ -19,6 +23,12 @@ function generateId(): string {
 
 export interface PollLoopConfig {
   provider: AgentProvider;
+  /**
+   * Name of the provider (e.g. "claude", "codex", "opencode"). Used to key
+   * the stored continuation per-provider so flipping providers doesn't
+   * resurrect a stale id from a different backend.
+   */
+  providerName: string;
   cwd: string;
   systemContext?: {
     instructions?: string;
@@ -39,8 +49,9 @@ export async function runPollLoop(config: PollLoopConfig): Promise<void> {
   // Resume the agent's prior session from a previous container run if one
   // was persisted. The continuation is opaque to the poll-loop — the
   // provider decides how to use it (Claude resumes a .jsonl transcript,
-  // other providers may reload a thread ID, etc.).
-  let continuation: string | undefined = getStoredSessionId();
+  // other providers may reload a thread ID, etc.). Keyed per-provider so
+  // a Codex thread id never gets handed to Claude or vice versa.
+  let continuation: string | undefined = migrateLegacyContinuation(config.providerName);
 
   if (continuation) {
     log(`Resuming agent session ${continuation}`);
@@ -94,7 +105,7 @@ export async function runPollLoop(config: PollLoopConfig): Promise<void> {
       if ((msg.kind === 'chat' || msg.kind === 'chat-sdk') && isClearCommand(msg)) {
         log('Clearing session (resetting continuation)');
         continuation = undefined;
-        clearStoredSessionId();
+        clearContinuation(config.providerName);
         writeMessageOut({
           id: generateId(),
           kind: 'chat',
@@ -160,10 +171,10 @@ export async function runPollLoop(config: PollLoopConfig): Promise<void> {
     const skippedSet = new Set(skipped);
     const processingIds = ids.filter((id) => !commandIds.includes(id) && !skippedSet.has(id));
     try {
-      const result = await processQuery(query, routing, processingIds);
+      const result = await processQuery(query, routing, processingIds, config.providerName);
       if (result.continuation && result.continuation !== continuation) {
         continuation = result.continuation;
-        setStoredSessionId(continuation);
+        setContinuation(config.providerName, continuation);
       }
     } catch (err) {
       const errMsg = err instanceof Error ? err.message : String(err);
@@ -175,7 +186,7 @@ export async function runPollLoop(config: PollLoopConfig): Promise<void> {
       if (continuation && config.provider.isSessionInvalid(err)) {
         log(`Stale session detected (${continuation}) — clearing for next retry`);
         continuation = undefined;
-        clearStoredSessionId();
+        clearContinuation(config.providerName);
       }
 
       // Write error response so the user knows something went wrong
@@ -238,6 +249,7 @@ async function processQuery(
   query: AgentQuery,
   routing: RoutingContext,
   initialBatchIds: string[],
+  providerName: string,
 ): Promise<QueryResult> {
   let queryContinuation: string | undefined;
   let done = false;
@@ -288,7 +300,7 @@ async function processQuery(
         // container died between `init` and `result`, the SDK session was
         // effectively orphaned and the next message started a blank
         // Claude session with no prior context.
-        setStoredSessionId(event.continuation);
+        setContinuation(providerName, event.continuation);
       } else if (event.type === 'result') {
         // A result — with or without text — means the turn is done. Mark
         // the initial batch completed now so the host sweep doesn't see

container/agent-runner/src/providers/codex-app-server.test.ts

@@ -0,0 +1,162 @@
+import { describe, expect, it, afterEach } from 'bun:test';
+import fs from 'fs';
+import os from 'os';
+import path from 'path';
+
+import {
+  type AppServer,
+  attachCodexAutoApproval,
+  buildCodexProcessEnv,
+  tomlBasicString,
+  writeCodexConfigToml,
+} from './codex-app-server.js';
+
+let tmpHome: string | null = null;
+const originalHome = process.env.HOME;
+
+afterEach(() => {
+  process.env.HOME = originalHome;
+  if (tmpHome) {
+    fs.rmSync(tmpHome, { recursive: true, force: true });
+    tmpHome = null;
+  }
+});
+
+describe('Codex config TOML', () => {
+  it('escapes basic strings', () => {
+    expect(tomlBasicString('a "quoted" \\\\ value')).toBe('"a \\"quoted\\" \\\\\\\\ value"');
+  });
+
+  it('rejects newlines', () => {
+    expect(() => tomlBasicString('bad\nvalue')).toThrow(/newline/);
+  });
+
+  it('hardcodes danger-full-access + never and writes model, effort, and MCP servers', () => {
+    tmpHome = fs.mkdtempSync(path.join(os.tmpdir(), 'codex-home-'));
+    process.env.HOME = tmpHome;
+
+    writeCodexConfigToml(
+      {
+        nanoclaw: {
+          command: 'bun',
+          args: ['run', '/app/src/mcp-tools/index.ts'],
+          env: { FOO: 'bar' },
+        },
+      },
+      { model: 'gpt-5', effort: 'medium' },
+    );
+
+    const content = fs.readFileSync(path.join(tmpHome, '.codex', 'config.toml'), 'utf-8');
+    expect(content).toContain('sandbox_mode = "danger-full-access"');
+    expect(content).toContain('approval_policy = "never"');
+    expect(content).toContain('project_doc_max_bytes = 32768');
+    expect(content).toContain('model = "gpt-5"');
+    expect(content).toContain('model_reasoning_effort = "medium"');
+    expect(content).not.toContain('[sandbox_workspace_write]');
+    expect(content).not.toContain('writable_roots =');
+    expect(content).toContain('[mcp_servers.nanoclaw]');
+    expect(content).toContain('command = "bun"');
+    expect(content).toContain('args = ["run", "/app/src/mcp-tools/index.ts"]');
+    expect(content).toContain('[mcp_servers.nanoclaw.env]');
+    expect(content).toContain('FOO = "bar"');
+  });
+});
+
+describe('Codex auto-approval', () => {
+  // NanoClaw (container isolation + OneCLI) is the boundary, so the handler accepts
+  // every request unconditionally — even paths/commands a sandbox policy would refuse.
+  it('grants full filesystem + network for permission requests', () => {
+    const { server, writes } = fakeServer();
+    attachCodexAutoApproval(server);
+
+    server.serverRequestHandlers[0]({
+      id: 1,
+      method: 'item/permissions/requestApproval',
+      params: { permissions: { fileSystem: { read: ['/workspace/agent'], write: ['/workspace/agent'] } } },
+    });
+
+    const result = JSON.parse(writes[0]).result as {
+      permissions: { fileSystem: { read: string[]; write: string[] }; network: { enabled: boolean } };
+      scope: string;
+    };
+    expect(result.scope).toBe('turn');
+    expect(result.permissions.fileSystem.read).toEqual(['/']);
+    expect(result.permissions.fileSystem.write).toEqual(['/']);
+    expect(result.permissions.network.enabled).toBe(true);
+  });
+
+  it('accepts file-change and command-exec approvals regardless of path', () => {
+    const { server, writes } = fakeServer();
+    attachCodexAutoApproval(server);
+
+    server.serverRequestHandlers[0]({ id: 2, method: 'item/fileChange/requestApproval', params: { grantRoot: '/etc' } });
+    server.serverRequestHandlers[0]({
+      id: 3,
+      method: 'item/commandExecution/requestApproval',
+      params: { command: 'rm -rf /', cwd: '/' },
+    });
+
+    expect(JSON.parse(writes[0]).result).toEqual({ decision: 'accept' });
+    expect(JSON.parse(writes[1]).result).toEqual({ decision: 'accept' });
+  });
+
+  it('approves legacy patch and command-exec approvals regardless of path', () => {
+    const { server, writes } = fakeServer();
+    attachCodexAutoApproval(server);
+
+    server.serverRequestHandlers[0]({
+      id: 4,
+      method: 'applyPatchApproval',
+      params: { fileChanges: { '/etc/passwd': {} } },
+    });
+    server.serverRequestHandlers[0]({ id: 5, method: 'execCommandApproval', params: { command: 'rm -rf /', cwd: '/' } });
+
+    expect(JSON.parse(writes[0]).result).toEqual({ decision: 'approved' });
+    expect(JSON.parse(writes[1]).result).toEqual({ decision: 'approved' });
+  });
+
+  it('fails closed for unknown server requests', () => {
+    const { server, writes } = fakeServer();
+    attachCodexAutoApproval(server);
+
+    server.serverRequestHandlers[0]({ id: 6, method: 'new/unknown/request' });
+
+    const response = JSON.parse(writes[0]);
+    expect(response.error.message).toContain('Unhandled Codex app-server request');
+  });
+});
+
+describe('Codex process env', () => {
+  it('forwards proxy/runtime env without leaking secret-like host env', () => {
+    const env = buildCodexProcessEnv({
+      PATH: '/bin',
+      HOME: '/home/node',
+      CODEX_HOME: '/home/node/.codex',
+      HTTPS_PROXY: 'http://proxy',
+      OPENAI_API_KEY: 'sk-test',
+      ONECLI_API_KEY: 'onecli-secret',
+      SOME_TOKEN: 'token',
+    });
+
+    expect(env.PATH).toBe('/bin');
+    expect(env.HOME).toBe('/home/node');
+    expect(env.CODEX_HOME).toBe('/home/node/.codex');
+    expect(env.HTTPS_PROXY).toBe('http://proxy');
+    expect(env.OPENAI_API_KEY).toBeUndefined();
+    expect(env.ONECLI_API_KEY).toBeUndefined();
+    expect(env.SOME_TOKEN).toBeUndefined();
+  });
+});
+
+function fakeServer(): { server: AppServer; writes: string[] } {
+  const writes: string[] = [];
+  const server = {
+    process: { stdin: { write: (line: string) => writes.push(line) } },
+    readline: { close: () => {} },
+    pending: new Map(),
+    notificationHandlers: [],
+    exitHandlers: [],
+    serverRequestHandlers: [],
+  } as unknown as AppServer;
+  return { server, writes };
+}

container/agent-runner/src/providers/codex-app-server.ts

@@ -0,0 +1,441 @@
+import fs from 'fs';
+import path from 'path';
+import { spawn, type ChildProcess } from 'child_process';
+import { createInterface, type Interface as ReadlineInterface } from 'readline';
+
+// Cap Codex's project-doc loading (AGENTS.md). The host-side composer
+// (src/providers/codex-agents-md.ts) enforces the same cap at compose time —
+// host and container share no modules, so the constant lives in both.
+const CODEX_PROJECT_DOC_MAX_BYTES = 32 * 1024;
+
+function log(msg: string): void {
+  console.error(`[codex-app-server] ${msg}`);
+}
+
+const INIT_TIMEOUT_MS = 30_000;
+
+export const STALE_THREAD_RE = /thread\s+not\s+found|unknown\s+thread|thread[_\s]id|no such thread/i;
+
+let nextRequestId = 1;
+
+export interface JsonRpcResponse {
+  id: number | string;
+  result?: unknown;
+  error?: { code: number; message: string; data?: unknown };
+}
+
+export interface JsonRpcNotification {
+  method: string;
+  params?: Record<string, unknown>;
+}
+
+export interface JsonRpcServerRequest {
+  id: number | string;
+  method: string;
+  params?: Record<string, unknown>;
+}
+
+type JsonRpcMessage = JsonRpcResponse | JsonRpcNotification | JsonRpcServerRequest;
+
+export interface AppServer {
+  process: ChildProcess;
+  readline: ReadlineInterface;
+  pending: Map<number | string, { resolve: (r: JsonRpcResponse) => void; reject: (e: Error) => void }>;
+  notificationHandlers: Array<(n: JsonRpcNotification) => void>;
+  serverRequestHandlers: Array<(r: JsonRpcServerRequest) => void>;
+  /**
+   * Fired when the app-server process dies (exit or spawn error). Pending
+   * request/response pairs are rejected separately via failPending — but a
+   * turn in flight has NO pending request (turn/start already resolved); it
+   * is parked on a notification waker that a dead process will never kick.
+   * Without these handlers a mid-turn crash surfaces as a 10-minute turn
+   * timeout instead of the real exit code, after the --rm container has
+   * already taken the server's stderr with it.
+   */
+  exitHandlers: Array<(err: Error) => void>;
+}
+
+export interface CodexMcpServer {
+  command: string;
+  args?: string[];
+  env?: Record<string, string>;
+}
+
+export type CodexReasoningEffort = 'none' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh';
+
+// Codex runs unrestricted inside the container. NanoClaw's container isolation and
+// the OneCLI allow-list are the security boundary — not Codex's own sandbox/approval
+// primitives (which can't run here anyway: workspace-write/read-only need user
+// namespaces, which the agent containers deny). Both are hardcoded as instance-level
+// defaults in config.toml; threads and turns inherit them, never override them.
+const CODEX_SANDBOX_MODE = 'danger-full-access';
+const CODEX_APPROVAL_POLICY = 'never';
+
+const CODEX_ENV_ALLOWLIST = new Set([
+  'ALL_PROXY',
+  'CURL_CA_BUNDLE',
+  'GIT_SSL_CAINFO',
+  'HOME',
+  'HTTP_PROXY',
+  'HTTPS_PROXY',
+  'LANG',
+  'LC_ALL',
+  'NODE_EXTRA_CA_CERTS',
+  'NO_PROXY',
+  'PATH',
+  'PNPM_HOME',
+  'REQUESTS_CA_BUNDLE',
+  'SSL_CERT_DIR',
+  'SSL_CERT_FILE',
+  'TEMP',
+  'TERM',
+  'TMP',
+  'TMPDIR',
+  'TZ',
+  'USER',
+  'all_proxy',
+  'http_proxy',
+  'https_proxy',
+  'no_proxy',
+  'CODEX_HOME',
+]);
+
+export interface ThreadParams {
+  model?: string;
+  cwd: string;
+  baseInstructions?: string;
+  developerInstructions?: string;
+}
+
+export interface TurnParams {
+  threadId: string;
+  inputText: string;
+  model?: string;
+  effort?: string;
+  cwd?: string;
+}
+
+export function spawnCodexAppServer(): AppServer {
+  const args = ['app-server', '--listen', 'stdio://'];
+  log(`Spawning: codex ${args.join(' ')}`);
+
+  const proc = spawn('codex', args, {
+    stdio: ['pipe', 'pipe', 'pipe'],
+    env: buildCodexProcessEnv(process.env),
+  });
+  const rl = createInterface({ input: proc.stdout! });
+
+  const server: AppServer = {
+    process: proc,
+    readline: rl,
+    pending: new Map(),
+    notificationHandlers: [],
+    exitHandlers: [],
+    serverRequestHandlers: [],
+  };
+
+  proc.stderr?.on('data', (chunk: Buffer) => {
+    const text = chunk.toString().trim();
+    if (text) log(`[stderr] ${text}`);
+  });
+
+  rl.on('line', (line: string) => {
+    if (!line.trim()) return;
+    let msg: JsonRpcMessage;
+    try {
+      msg = JSON.parse(line) as JsonRpcMessage;
+    } catch {
+      log(`[parse-error] ${line.slice(0, 200)}`);
+      return;
+    }
+
+    if (isResponse(msg)) {
+      const handler = server.pending.get(msg.id);
+      if (handler) {
+        server.pending.delete(msg.id);
+        handler.resolve(msg);
+      }
+    } else if (isServerRequest(msg)) {
+      for (const h of server.serverRequestHandlers) h(msg);
+    } else if ('method' in msg) {
+      for (const h of server.notificationHandlers) h(msg as JsonRpcNotification);
+    }
+  });
+
+  const failPending = (err: Error): void => {
+    for (const [, handler] of server.pending) handler.reject(err);
+    server.pending.clear();
+  };
+
+  proc.on('error', (err) => {
+    log(`[process-error] ${err.message}`);
+    failPending(err);
+    for (const h of [...server.exitHandlers]) h(err);
+  });
+
+  proc.on('exit', (code, signal) => {
+    log(`[exit] code=${code} signal=${signal}`);
+    const err = new Error(`Codex app-server exited: code=${code} signal=${signal}`);
+    failPending(err);
+    for (const h of [...server.exitHandlers]) h(err);
+  });
+
+  return server;
+}
+
+export function sendCodexRequest(
+  server: AppServer,
+  method: string,
+  params?: Record<string, unknown>,
+  timeoutMs = 60_000,
+): Promise<JsonRpcResponse> {
+  const id = nextRequestId++;
+  const req = params === undefined ? { id, method } : { id, method, params };
+  const line = JSON.stringify(req) + '\n';
+
+  return new Promise<JsonRpcResponse>((resolve, reject) => {
+    const timer = setTimeout(() => {
+      server.pending.delete(id);
+      reject(new Error(`Timeout waiting for ${method} response (${timeoutMs}ms)`));
+    }, timeoutMs);
+
+    server.pending.set(id, {
+      resolve: (r) => {
+        clearTimeout(timer);
+        resolve(r);
+      },
+      reject: (e) => {
+        clearTimeout(timer);
+        reject(e);
+      },
+    });
+
+    try {
+      server.process.stdin!.write(line);
+    } catch (err) {
+      clearTimeout(timer);
+      server.pending.delete(id);
+      reject(err instanceof Error ? err : new Error(String(err)));
+    }
+  });
+}
+
+export function sendCodexNotification(server: AppServer, method: string, params?: Record<string, unknown>): void {
+  const line = JSON.stringify(params === undefined ? { method } : { method, params }) + '\n';
+  server.process.stdin!.write(line);
+}
+
+export function sendCodexResponse(server: AppServer, id: number | string, result: unknown): void {
+  try {
+    server.process.stdin!.write(JSON.stringify({ id, result }) + '\n');
+  } catch (err) {
+    log(`[send-error] response id=${id}: ${err instanceof Error ? err.message : String(err)}`);
+  }
+}
+
+export function killCodexAppServer(server: AppServer): void {
+  try {
+    server.readline.close();
+    server.process.kill('SIGTERM');
+  } catch {
+    /* ignore */
+  }
+}
+
+export async function initializeCodexAppServer(server: AppServer): Promise<void> {
+  const resp = await sendCodexRequest(
+    server,
+    'initialize',
+    {
+      clientInfo: { name: 'nanoclaw', title: 'NanoClaw', version: '2.0' },
+      capabilities: { experimentalApi: true },
+    },
+    INIT_TIMEOUT_MS,
+  );
+  if (resp.error) throw new Error(`initialize failed: ${resp.error.message}`);
+  sendCodexNotification(server, 'initialized');
+}
+
+export async function startOrResumeCodexThread(
+  server: AppServer,
+  threadId: string | undefined,
+  params: ThreadParams,
+): Promise<string> {
+  const baseParams = {
+    model: params.model,
+    cwd: params.cwd,
+    approvalPolicy: CODEX_APPROVAL_POLICY,
+    sandbox: CODEX_SANDBOX_MODE,
+    baseInstructions: params.baseInstructions,
+    developerInstructions: params.developerInstructions,
+    personality: 'friendly',
+    sessionStartSource: 'startup',
+    persistExtendedHistory: false,
+  };
+
+  if (threadId) {
+    const resp = await sendCodexRequest(server, 'thread/resume', {
+      threadId,
+      ...baseParams,
+      excludeTurns: true,
+    });
+    if (!resp.error) return threadId;
+    if (!STALE_THREAD_RE.test(resp.error.message)) {
+      throw new Error(`thread/resume failed: ${resp.error.message}`);
+    }
+    log(`Stale thread ${threadId}; starting fresh thread.`);
+  }
+
+  const resp = await sendCodexRequest(server, 'thread/start', {
+    ...baseParams,
+    experimentalRawEvents: false,
+  });
+  if (resp.error) throw new Error(`thread/start failed: ${resp.error.message}`);
+
+  const result = resp.result as { thread?: { id?: string } } | undefined;
+  const newThreadId = result?.thread?.id;
+  if (!newThreadId) throw new Error('thread/start response missing thread ID');
+  return newThreadId;
+}
+
+export async function startCodexTurn(server: AppServer, params: TurnParams): Promise<string> {
+  const resp = await sendCodexRequest(server, 'turn/start', {
+    threadId: params.threadId,
+    input: [{ type: 'text', text: params.inputText, text_elements: [] }],
+    model: params.model,
+    effort: params.effort,
+    cwd: params.cwd,
+  });
+  if (resp.error) throw new Error(`turn/start failed: ${resp.error.message}`);
+  const result = resp.result as { turn?: { id?: string } } | undefined;
+  const turnId = result?.turn?.id;
+  if (!turnId) throw new Error('turn/start response missing turn ID');
+  return turnId;
+}
+
+export async function steerCodexTurn(
+  server: AppServer,
+  threadId: string,
+  turnId: string,
+  inputText: string,
+): Promise<void> {
+  const resp = await sendCodexRequest(server, 'turn/steer', {
+    threadId,
+    expectedTurnId: turnId,
+    input: [{ type: 'text', text: inputText, text_elements: [] }],
+  });
+  if (resp.error) throw new Error(`turn/steer failed: ${resp.error.message}`);
+}
+
+export async function interruptCodexTurn(server: AppServer, threadId: string, turnId: string): Promise<void> {
+  const resp = await sendCodexRequest(server, 'turn/interrupt', { threadId, turnId }, 10_000);
+  if (resp.error) throw new Error(`turn/interrupt failed: ${resp.error.message}`);
+}
+
+// With approval_policy=never the command/patch approval requests don't fire, but the
+// app-server still sends a few non-approval server→client requests (permission
+// negotiation, MCP elicitations, tool calls) that must be answered or the turn hangs.
+// NanoClaw is the boundary, so accept/grant everything.
+export function attachCodexAutoApproval(server: AppServer): void {
+  server.serverRequestHandlers.push((req) => {
+    switch (req.method) {
+      case 'item/commandExecution/requestApproval':
+      case 'item/fileChange/requestApproval':
+        sendCodexResponse(server, req.id, { decision: 'accept' });
+        break;
+      case 'applyPatchApproval':
+      case 'execCommandApproval':
+        sendCodexResponse(server, req.id, { decision: 'approved' });
+        break;
+      case 'item/permissions/requestApproval':
+        sendCodexResponse(server, req.id, {
+          permissions: { fileSystem: { read: ['/'], write: ['/'] }, network: { enabled: true } },
+          scope: 'turn',
+          strictAutoReview: true,
+        });
+        break;
+      case 'item/tool/requestUserInput':
+        sendCodexResponse(server, req.id, { answers: {} });
+        break;
+      case 'mcpServer/elicitation/request':
+        sendCodexResponse(server, req.id, { action: 'cancel', content: null, _meta: null });
+        break;
+      case 'item/tool/call':
+        sendCodexResponse(server, req.id, { success: false, contentItems: [] });
+        break;
+      default:
+        sendCodexError(server, req.id, `Unhandled Codex app-server request: ${req.method}`);
+        break;
+    }
+  });
+}
+
+export function writeCodexConfigToml(
+  servers: Record<string, CodexMcpServer>,
+  opts: { model?: string; effort?: string } = {},
+): void {
+  const codexConfigDir = path.join(process.env.HOME || '/home/node', '.codex');
+  fs.mkdirSync(codexConfigDir, { recursive: true });
+  const configTomlPath = path.join(codexConfigDir, 'config.toml');
+
+  // Instance-level defaults the app-server reads on startup; threads/turns inherit them.
+  const lines: string[] = [
+    `sandbox_mode = ${tomlBasicString(CODEX_SANDBOX_MODE)}`,
+    `approval_policy = ${tomlBasicString(CODEX_APPROVAL_POLICY)}`,
+    `project_doc_max_bytes = ${CODEX_PROJECT_DOC_MAX_BYTES}`,
+  ];
+  if (opts.model) lines.push(`model = ${tomlBasicString(opts.model)}`);
+  if (opts.effort) lines.push(`model_reasoning_effort = ${tomlBasicString(opts.effort)}`);
+  lines.push('');
+
+  for (const [name, config] of Object.entries(servers)) {
+    lines.push(`[mcp_servers.${name}]`);
+    lines.push(`command = ${tomlBasicString(config.command)}`);
+    if (config.args && config.args.length > 0) {
+      lines.push(`args = [${config.args.map(tomlBasicString).join(', ')}]`);
+    }
+    if (config.env && Object.keys(config.env).length > 0) {
+      lines.push(`[mcp_servers.${name}.env]`);
+      for (const [key, value] of Object.entries(config.env)) {
+        lines.push(`${key} = ${tomlBasicString(value)}`);
+      }
+    }
+    lines.push('');
+  }
+
+  fs.writeFileSync(configTomlPath, lines.join('\n'));
+}
+
+export function buildCodexProcessEnv(env: NodeJS.ProcessEnv): NodeJS.ProcessEnv {
+  const next: NodeJS.ProcessEnv = {};
+  for (const key of CODEX_ENV_ALLOWLIST) {
+    const value = env[key];
+    if (value !== undefined) next[key] = value;
+  }
+  if (!next.CODEX_HOME) next.CODEX_HOME = next.HOME ? path.join(next.HOME, '.codex') : '/home/node/.codex';
+  if (!next.HOME) next.HOME = '/home/node';
+  return next;
+}
+
+export function tomlBasicString(value: string): string {
+  if (value.includes('\n') || value.includes('\r')) {
+    throw new Error(`MCP config value contains newline: ${JSON.stringify(value.slice(0, 40))}`);
+  }
+  return `"${value.replace(/\\/g, '\\\\').replace(/"/g, '\\"')}"`;
+}
+
+function sendCodexError(server: AppServer, id: number | string, message: string, data?: unknown): void {
+  try {
+    server.process.stdin!.write(JSON.stringify({ id, error: { code: -32000, message, data } }) + '\n');
+  } catch (err) {
+    log(`[send-error] error id=${id}: ${err instanceof Error ? err.message : String(err)}`);
+  }
+}
+
+function isResponse(msg: JsonRpcMessage): msg is JsonRpcResponse {
+  return 'id' in msg && ('result' in msg || 'error' in msg) && !('method' in msg);
+}
+
+function isServerRequest(msg: JsonRpcMessage): msg is JsonRpcServerRequest {
+  return 'id' in msg && 'method' in msg;
+}

container/agent-runner/src/providers/codex-cli-tools.test.ts

@@ -0,0 +1,39 @@
+// Structural guard for the Codex CLI install in container/cli-tools.json.
+//
+// @openai/codex is a CLI *binary* installed from the global-CLI manifest (a
+// json-merge seam), not an importable package, so the barrel-driven
+// registration tests cannot see it. This test reads the real cli-tools.json
+// and asserts the @openai/codex entry is present and pinned to an exact
+// version. It goes red if the manifest entry is dropped or unpins.
+//
+// Runs under bun (same suite as the container registration test):
+//   cd container/agent-runner && bun test src/providers/codex-cli-tools.test.ts
+
+import { existsSync, readFileSync } from 'fs';
+import path from 'path';
+
+import { describe, it, expect } from 'bun:test';
+
+// container/agent-runner/src/providers/ -> container/cli-tools.json
+const MANIFEST = path.join(import.meta.dir, '..', '..', '..', 'cli-tools.json');
+const manifestPresent = existsSync(MANIFEST);
+
+// Read lazily — `describe.skipIf` still runs the body to register tests, so the
+// read has to be guarded for the bare-branch (no manifest) case.
+const tools: Array<{ name: string; version: string }> = manifestPresent
+  ? JSON.parse(readFileSync(MANIFEST, 'utf8'))
+  : [];
+const codex = tools.find((t) => t.name === '@openai/codex');
+
+// cli-tools.json is a trunk file; on the bare providers branch it isn't present,
+// so skip there. In an installed tree (trunk + this payload) it must carry the
+// pinned @openai/codex entry.
+describe.skipIf(!manifestPresent)('container/cli-tools.json codex CLI install', () => {
+  it('includes the @openai/codex entry', () => {
+    expect(codex).toBeDefined();
+  });
+
+  it('pins it to an exact semver (no latest, no ranges)', () => {
+    expect(codex?.version).toMatch(/^\d+\.\d+\.\d+(?:[-+][0-9A-Za-z.-]+)?$/);
+  });
+});

container/agent-runner/src/providers/codex-registration.test.ts

@@ -0,0 +1,22 @@
+/**
+ * Integration test for the codex provider's CONTAINER-side reach-in: the self-registration
+ * import in container/agent-runner/src/providers/index.ts. Importing the barrel runs
+ * codex.ts's top-level registerProvider('codex', …); without that import line
+ * createProvider('codex') throws 'Unknown provider' at runtime.
+ *
+ * Behavior, not structural, and BARREL-ONLY: it imports the real barrel (./index.js),
+ * never ./codex.js directly, then asserts listProviderNames() contains the provider. The
+ * existing codex.factory.test.ts imports ./codex.js directly, so it self-registers and
+ * stays GREEN when the barrel line is deleted — a unit test, not a registration guard.
+ * This goes red if the barrel import is deleted/drifts or the barrel fails to evaluate. codex uses the @openai/codex CLI *binary* (not an importable package), so this test does not guard that dependency — the Dockerfile install line is guarded structurally + by the container build (see the skill validate step).
+ */
+import { describe, it, expect } from 'bun:test';
+
+import { listProviderNames } from './provider-registry.js';
+import './index.js'; // the real container provider barrel — triggers each provider's registerProvider()
+
+describe('codex provider registration', () => {
+  it('registers codex via the provider barrel', () => {
+    expect(listProviderNames()).toContain('codex');
+  });
+});

container/agent-runner/src/providers/codex.factory.test.ts

@@ -0,0 +1,17 @@
+import { describe, expect, it } from 'bun:test';
+
+import { CodexProvider } from './codex.js';
+
+describe('CodexProvider', () => {
+  it('rejects unsupported reasoning effort values', () => {
+    expect(() => new CodexProvider({ effort: 'max' })).toThrow(/Unsupported Codex reasoning effort/);
+  });
+
+  it('normalizes supported reasoning effort values', () => {
+    expect(new CodexProvider({ effort: 'HIGH' })).toBeInstanceOf(CodexProvider);
+  });
+
+  it('accepts supported reasoning effort values', () => {
+    expect(new CodexProvider({ effort: 'xhigh' })).toBeInstanceOf(CodexProvider);
+  });
+});

container/agent-runner/src/providers/codex.ts

@@ -0,0 +1,419 @@
+import fs from 'fs';
+import path from 'path';
+
+import { registerProvider } from './provider-registry.js';
+import type {
+  AgentProvider,
+  AgentQuery,
+  McpServerConfig,
+  ProviderEvent,
+  ProviderExchange,
+  ProviderOptions,
+  QueryInput,
+} from './types.js';
+import { archiveProviderExchange } from './exchange-archive.js';
+import {
+  type AppServer,
+  type CodexReasoningEffort,
+  type JsonRpcNotification,
+  STALE_THREAD_RE,
+  attachCodexAutoApproval,
+  initializeCodexAppServer,
+  interruptCodexTurn,
+  killCodexAppServer,
+  spawnCodexAppServer,
+  startCodexTurn,
+  startOrResumeCodexThread,
+  steerCodexTurn,
+  writeCodexConfigToml,
+} from './codex-app-server.js';
+
+const TURN_TIMEOUT_MS = 10 * 60 * 1000;
+const SUPPORTED_EFFORTS = new Set<CodexReasoningEffort>(['none', 'minimal', 'low', 'medium', 'high', 'xhigh']);
+
+export interface CodexRuntimeDeps {
+  writeCodexConfigToml: typeof writeCodexConfigToml;
+  spawnCodexAppServer: typeof spawnCodexAppServer;
+  attachCodexAutoApproval: typeof attachCodexAutoApproval;
+  initializeCodexAppServer: typeof initializeCodexAppServer;
+  startOrResumeCodexThread: typeof startOrResumeCodexThread;
+  startCodexTurn: typeof startCodexTurn;
+  steerCodexTurn: typeof steerCodexTurn;
+  interruptCodexTurn: typeof interruptCodexTurn;
+  killCodexAppServer: typeof killCodexAppServer;
+}
+
+const defaultCodexRuntimeDeps: CodexRuntimeDeps = {
+  writeCodexConfigToml,
+  spawnCodexAppServer,
+  attachCodexAutoApproval,
+  initializeCodexAppServer,
+  startOrResumeCodexThread,
+  startCodexTurn,
+  steerCodexTurn,
+  interruptCodexTurn,
+  killCodexAppServer,
+};
+
+function classifyError(message: string): string | undefined {
+  if (/auth|api key|unauthorized|login|credential/i.test(message)) return 'auth';
+  if (/quota|rate limit|insufficient|billing|credit/i.test(message)) return 'quota';
+  if (/sandbox|permission|denied/i.test(message)) return 'sandbox';
+  if (/thread|conversation|session/i.test(message)) return 'stale-session';
+  return undefined;
+}
+
+function normalizeEffort(effort: string | undefined): CodexReasoningEffort | undefined {
+  const normalized = effort?.trim().toLowerCase();
+  if (!normalized) return undefined;
+  if (!SUPPORTED_EFFORTS.has(normalized as CodexReasoningEffort)) {
+    throw new Error(`Unsupported Codex reasoning effort: ${effort}`);
+  }
+  return normalized as CodexReasoningEffort;
+}
+
+export class CodexProvider implements AgentProvider {
+  readonly supportsNativeSlashCommands = false;
+  // Codex has no native NanoClaw memory — opt in to the runner's persistent
+  // memory/ scaffold (see memory-scaffold.ts).
+  readonly usesMemoryScaffold = true;
+  // The app-server keeps history server-side; there is no on-disk transcript,
+  // so the provider persists each exchange itself into `conversations/`
+  // (see exchange-archive.ts). The poll-loop reports exchanges through this
+  // hook and does nothing else — archiving is payload code, not runner code.
+  onExchangeComplete(exchange: ProviderExchange): void {
+    archiveProviderExchange({
+      provider: 'codex',
+      prompt: exchange.prompt,
+      result: exchange.result,
+      continuation: exchange.continuation,
+      status: exchange.status,
+    });
+  }
+
+  private readonly mcpServers: Record<string, McpServerConfig>;
+  private readonly model?: string;
+  private readonly effort?: CodexReasoningEffort;
+  private readonly runtime: CodexRuntimeDeps;
+
+  constructor(options: ProviderOptions = {}, runtime: CodexRuntimeDeps = defaultCodexRuntimeDeps) {
+    this.mcpServers = options.mcpServers ?? {};
+    this.model = options.model;
+    this.runtime = runtime;
+    this.effort = normalizeEffort(options.effort);
+  }
+
+  isSessionInvalid(err: unknown): boolean {
+    const msg = err instanceof Error ? err.message : String(err);
+    return STALE_THREAD_RE.test(msg);
+  }
+
+  query(input: QueryInput): AgentQuery {
+    const pending: string[] = [input.prompt];
+    let waiting: (() => void) | null = null;
+    let ended = false;
+    let aborted = false;
+    let activeServer: AppServer | null = null;
+    let activeThreadId: string | null = null;
+    let activeTurnId: string | null = null;
+    let wakeActiveTurn: (() => void) | null = null;
+
+    const wake = (): void => {
+      waiting?.();
+      waiting = null;
+    };
+
+    const pushOrSteer = (message: string): void => {
+      if (activeServer && activeThreadId && activeTurnId) {
+        void this.runtime.steerCodexTurn(activeServer, activeThreadId, activeTurnId, message).catch(() => {
+          pending.push(message);
+          wake();
+        });
+        return;
+      }
+      pending.push(message);
+      wake();
+    };
+
+    const self = this;
+
+    async function* gen(): AsyncGenerator<ProviderEvent> {
+      self.runtime.writeCodexConfigToml(self.mcpServers, { model: self.model, effort: self.effort });
+      const server = self.runtime.spawnCodexAppServer();
+      activeServer = server;
+      self.runtime.attachCodexAutoApproval(server);
+
+      let threadId: string | undefined = input.continuation;
+      let initYielded = false;
+
+      try {
+        await self.runtime.initializeCodexAppServer(server);
+        threadId = await self.runtime.startOrResumeCodexThread(server, threadId, {
+          model: self.model,
+          cwd: input.cwd,
+          baseInstructions: input.systemContext?.instructions,
+        });
+        activeThreadId = threadId;
+
+        while (!aborted) {
+          while (pending.length === 0 && !ended && !aborted) {
+            await new Promise<void>((resolve) => {
+              waiting = resolve;
+            });
+          }
+          if (aborted) return;
+          if (pending.length === 0 && ended) return;
+
+          const text = pending.shift()!;
+          yield* runOneTurn(
+            server,
+            threadId,
+            text,
+            self.model,
+            self.effort,
+            input.cwd,
+            (turnId) => {
+              activeTurnId = turnId;
+            },
+            () => {
+              activeTurnId = null;
+            },
+            () => initYielded,
+            () => {
+              initYielded = true;
+            },
+            () => aborted,
+            (waker) => {
+              wakeActiveTurn = waker;
+            },
+            self.runtime.startCodexTurn,
+          );
+        }
+      } finally {
+        activeTurnId = null;
+        activeThreadId = null;
+        activeServer = null;
+        wakeActiveTurn = null;
+        self.runtime.killCodexAppServer(server);
+      }
+    }
+
+    return {
+      push: pushOrSteer,
+      end: () => {
+        ended = true;
+        wake();
+      },
+      abort: () => {
+        aborted = true;
+        if (activeServer && activeThreadId && activeTurnId) {
+          void this.runtime.interruptCodexTurn(activeServer, activeThreadId, activeTurnId).catch(() => {});
+        }
+        wakeActiveTurn?.();
+        wake();
+      },
+      events: gen(),
+    };
+  }
+}
+
+async function* runOneTurn(
+  server: AppServer,
+  threadId: string,
+  inputText: string,
+  model: string | undefined,
+  effort: string | undefined,
+  cwd: string,
+  setActiveTurn: (turnId: string) => void,
+  clearActiveTurn: () => void,
+  hasInit: () => boolean,
+  markInit: () => void,
+  isAborted: () => boolean,
+  setAbortWaker: (waker: (() => void) | null) => void,
+  startTurn: typeof startCodexTurn,
+): AsyncGenerator<ProviderEvent> {
+  const state: { error: Error | null } = { error: null };
+  let resultText = '';
+  let turnDone = false;
+  let turnId: string | null = null;
+
+  // A finished turn can no longer absorb steered input: codex's turn/steer
+  // against a completed turn resolves as a no-op, so a follow-up routed there
+  // is lost silently. Clear the active-turn marker the moment the turn ends —
+  // before the generator drains and tears down in its `finally` — so
+  // pushOrSteer queues any racing follow-up into a fresh turn instead.
+  const finishTurn = (): void => {
+    turnDone = true;
+    clearActiveTurn();
+  };
+
+  const buffer: ProviderEvent[] = [];
+  let waker: (() => void) | null = null;
+  const kick = (): void => {
+    waker?.();
+    waker = null;
+  };
+  setAbortWaker(kick);
+
+  const handler = (n: JsonRpcNotification): void => {
+    const method = n.method;
+    const params = n.params ?? {};
+    buffer.push({ type: 'activity' });
+
+    switch (method) {
+      case 'thread/started': {
+        const thread = params.thread as { id?: string } | undefined;
+        if (thread?.id && !hasInit()) {
+          markInit();
+          buffer.push({ type: 'init', continuation: thread.id });
+        }
+        break;
+      }
+      case 'turn/started': {
+        const turn = params.turn as { id?: string } | undefined;
+        if (turn?.id) {
+          turnId = turn.id;
+          setActiveTurn(turn.id);
+        }
+        break;
+      }
+      case 'item/agentMessage/delta': {
+        const delta = params.delta as string | undefined;
+        if (delta) resultText += delta;
+        break;
+      }
+      case 'item/completed': {
+        const item = params.item as { type?: string; text?: string } | undefined;
+        if (item?.type === 'agentMessage' && item.text) resultText = item.text;
+        break;
+      }
+      case 'thread/status/changed': {
+        const status = params.status as string | undefined;
+        if (status) buffer.push({ type: 'progress', message: `status: ${status}` });
+        break;
+      }
+      case 'error': {
+        const err = params.error as { message?: string; additionalDetails?: string | null } | undefined;
+        const msg = [err?.message, err?.additionalDetails].filter(Boolean).join(': ') || 'Codex turn failed';
+        state.error = new Error(msg);
+        finishTurn();
+        break;
+      }
+      case 'turn/completed': {
+        const turn = params.turn as
+          | { error?: { message?: string; additionalDetails?: string | null } | null; items?: unknown[] }
+          | undefined;
+        const agentMessage = turn?.items
+          ?.filter((item): item is { type: string; text?: string } => typeof item === 'object' && item !== null)
+          .find((item) => item.type === 'agentMessage' && item.text);
+        if (agentMessage?.text) resultText = agentMessage.text;
+        if (turn?.error) {
+          const msg =
+            [turn.error.message, turn.error.additionalDetails].filter(Boolean).join(': ') || 'Codex turn failed';
+          state.error = new Error(msg);
+        }
+        finishTurn();
+        break;
+      }
+      default:
+        break;
+    }
+
+    kick();
+  };
+
+  server.notificationHandlers.push(handler);
+
+  // A dead app-server can't send the notification this turn is parked on —
+  // end the turn immediately with the real cause instead of the 10-min timeout.
+  const onServerExit = (err: Error): void => {
+    if (turnDone) return;
+    state.error = err;
+    finishTurn();
+    kick();
+  };
+  server.exitHandlers.push(onServerExit);
+
+  const timer = setTimeout(() => {
+    state.error = new Error(`Turn timed out after ${TURN_TIMEOUT_MS}ms`);
+    finishTurn();
+    kick();
+  }, TURN_TIMEOUT_MS);
+
+  try {
+    if (!hasInit()) {
+      markInit();
+      buffer.push({ type: 'init', continuation: threadId });
+    }
+
+    turnId = await startTurn(server, {
+      threadId,
+      inputText,
+      model,
+      effort,
+      cwd,
+    });
+    setActiveTurn(turnId);
+    const imagesBefore = listGeneratedImages(threadId);
+    if (isAborted()) return;
+
+    while (true) {
+      while (buffer.length > 0) {
+        yield buffer.shift()!;
+      }
+      if (turnDone || isAborted()) break;
+      await new Promise<void>((resolve) => {
+        waker = resolve;
+      });
+      waker = null;
+    }
+
+    while (buffer.length > 0) yield buffer.shift()!;
+
+    if (isAborted()) return;
+
+    if (state.error) {
+      yield {
+        type: 'error',
+        message: state.error.message,
+        retryable: false,
+        classification: classifyError(state.error.message),
+      };
+      throw state.error;
+    }
+
+    for (const imagePath of listGeneratedImages(threadId)) {
+      if (!imagesBefore.has(imagePath)) {
+        yield { type: 'file', path: imagePath };
+      }
+    }
+
+    yield { type: 'result', text: resultText || null };
+  } finally {
+    clearTimeout(timer);
+    clearActiveTurn();
+    setAbortWaker(null);
+    const idx = server.notificationHandlers.indexOf(handler);
+    if (idx >= 0) server.notificationHandlers.splice(idx, 1);
+    const exitIdx = server.exitHandlers.indexOf(onServerExit);
+    if (exitIdx >= 0) server.exitHandlers.splice(exitIdx, 1);
+  }
+}
+
+/**
+ * Codex's built-in image generation saves into CODEX_HOME/generated_images/
+ * <threadId>/ — its native client renders those to the user, so the model
+ * believes delivery already happened and won't send_file them. The runner
+ * must deliver them itself: snapshot the dir at turn start, emit a `file`
+ * event for anything new at turn end.
+ */
+function listGeneratedImages(threadId: string): Set<string> {
+  const dir = path.join(process.env.CODEX_HOME || '/home/node/.codex', 'generated_images', threadId);
+  try {
+    return new Set(fs.readdirSync(dir).map((f) => path.join(dir, f)));
+  } catch {
+    return new Set();
+  }
+}
+
+registerProvider('codex', (opts) => new CodexProvider(opts));

container/agent-runner/src/providers/codex.turns.test.ts

@@ -0,0 +1,267 @@
+import { afterEach, beforeEach, describe, expect, it } from 'bun:test';
+import fs from 'fs';
+import os from 'os';
+import path from 'path';
+
+import { CodexProvider, type CodexRuntimeDeps } from './codex.js';
+import type { AppServer, JsonRpcNotification, TurnParams } from './codex-app-server.js';
+import type { ProviderEvent } from './types.js';
+
+describe('CodexProvider active turns', () => {
+  it('steers follow-ups into the active turn and yields liveness activity', async () => {
+    const fake = createFakeCodexRuntime();
+    const provider = new CodexProvider({}, fake.runtime);
+    const query = provider.query({ prompt: 'first prompt', cwd: '/workspace/agent' });
+    const events: ProviderEvent[] = [];
+
+    const collect = collectEvents(query.events, events);
+
+    await waitFor(() => fake.startCalls.length === 1);
+    query.push('follow-up prompt');
+    await waitFor(() => fake.steerCalls.length === 1);
+    query.end();
+    fake.completeTurn('final answer');
+
+    await collect;
+
+    expect(fake.startCalls).toHaveLength(1);
+    expect(fake.startCalls[0].inputText).toBe('first prompt');
+    expect(fake.steerCalls).toEqual([{ threadId: 'thread-1', turnId: 'turn-1', inputText: 'follow-up prompt' }]);
+    expect(events.filter((event) => event.type === 'activity').length).toBeGreaterThanOrEqual(2);
+    expect(events.filter((event) => event.type === 'result')).toEqual([{ type: 'result', text: 'final answer' }]);
+    expect(fake.killed).toBe(true);
+  });
+
+  it('queues follow-ups for the next turn when steering is rejected', async () => {
+    const fake = createFakeCodexRuntime({ rejectSteer: true });
+    const provider = new CodexProvider({}, fake.runtime);
+    const query = provider.query({ prompt: 'first prompt', cwd: '/workspace/agent' });
+    const events: ProviderEvent[] = [];
+
+    const collect = collectEvents(query.events, events);
+
+    await waitFor(() => fake.startCalls.length === 1);
+    query.push('queued follow-up');
+    await waitFor(() => fake.steerCalls.length === 1);
+    await sleep(0);
+
+    fake.completeTurn('first answer');
+    await waitFor(() => fake.startCalls.length === 2);
+    query.end();
+    fake.completeTurn('second answer');
+
+    await collect;
+
+    expect(fake.startCalls.map((call) => call.inputText)).toEqual(['first prompt', 'queued follow-up']);
+    expect(fake.steerCalls).toHaveLength(1);
+    expect(events.filter((event) => event.type === 'result')).toEqual([
+      { type: 'result', text: 'first answer' },
+      { type: 'result', text: 'second answer' },
+    ]);
+  });
+
+  it('queues a follow-up that races turn completion into a new turn, never steering the finished turn', async () => {
+    const fake = createFakeCodexRuntime();
+    const provider = new CodexProvider({}, fake.runtime);
+    const query = provider.query({ prompt: 'first prompt', cwd: '/workspace/agent' });
+    const events: ProviderEvent[] = [];
+
+    const collect = collectEvents(query.events, events);
+
+    await waitFor(() => fake.startCalls.length === 1);
+
+    // The turn completes, then a follow-up lands in the same tick — before the
+    // generator has drained and torn the turn down. codex's turn/steer no-ops
+    // on a finished turn (resolves without error), so steering here would drop
+    // the message silently. It must start a fresh turn instead.
+    fake.completeTurn('first answer');
+    query.push('racing follow-up');
+
+    await waitFor(() => fake.startCalls.length === 2);
+    query.end();
+    fake.completeTurn('second answer');
+
+    await collect;
+
+    expect(fake.steerCalls).toHaveLength(0);
+    expect(fake.startCalls.map((call) => call.inputText)).toEqual(['first prompt', 'racing follow-up']);
+    expect(events.filter((event) => event.type === 'result')).toEqual([
+      { type: 'result', text: 'first answer' },
+      { type: 'result', text: 'second answer' },
+    ]);
+  });
+
+  it('interrupts the active turn and closes the stream on abort', async () => {
+    const fake = createFakeCodexRuntime();
+    const provider = new CodexProvider({}, fake.runtime);
+    const query = provider.query({ prompt: 'first prompt', cwd: '/workspace/agent' });
+    const events: ProviderEvent[] = [];
+
+    const collect = collectEvents(query.events, events);
+
+    await waitFor(() => fake.startCalls.length === 1);
+    query.abort();
+
+    await collect;
+
+    expect(fake.interruptCalls).toEqual([{ threadId: 'thread-1', turnId: 'turn-1' }]);
+    expect(events.some((event) => event.type === 'result')).toBe(false);
+    expect(fake.killed).toBe(true);
+  });
+
+  it('threads the configured model and effort into the turn', async () => {
+    const fake = createFakeCodexRuntime();
+    const provider = new CodexProvider({ model: 'gpt-5.5', effort: 'high' }, fake.runtime);
+    const query = provider.query({ prompt: 'first prompt', cwd: '/workspace/agent' });
+    const events: ProviderEvent[] = [];
+
+    const collect = collectEvents(query.events, events);
+
+    await waitFor(() => fake.startCalls.length === 1);
+    query.end();
+    fake.completeTurn('final answer');
+
+    await collect;
+
+    expect(fake.startCalls[0].model).toBe('gpt-5.5');
+    expect(fake.startCalls[0].effort).toBe('high');
+    expect(events.filter((event) => event.type === 'result')).toEqual([{ type: 'result', text: 'final answer' }]);
+  });
+
+  it('delivers harness-generated images as file events — the model never sends them itself', async () => {
+    const codexHome = fs.mkdtempSync(path.join(os.tmpdir(), 'codex-home-'));
+    const prevHome = process.env.CODEX_HOME;
+    process.env.CODEX_HOME = codexHome;
+    try {
+      const fake = createFakeCodexRuntime();
+      const provider = new CodexProvider({}, fake.runtime);
+      const query = provider.query({ prompt: 'make an image', cwd: '/workspace/agent' });
+      const events: ProviderEvent[] = [];
+      const collect = collectEvents(query.events, events);
+
+      await waitFor(() => fake.startCalls.length === 1);
+      // Codex's built-in image_gen writes into CODEX_HOME mid-turn.
+      const imagesDir = path.join(codexHome, 'generated_images', 'thread-1');
+      fs.mkdirSync(imagesDir, { recursive: true });
+      fs.writeFileSync(path.join(imagesDir, 'ig_abc.png'), 'png-bytes');
+
+      query.end();
+      fake.completeTurn('Here you go — created the image.');
+      await collect;
+
+      const files = events.filter((event) => event.type === 'file') as Array<{ type: 'file'; path: string }>;
+      expect(files).toHaveLength(1);
+      expect(files[0].path).toBe(path.join(imagesDir, 'ig_abc.png'));
+      // file events arrive before the result so delivery shares the turn.
+      expect(events.findIndex((e) => e.type === 'file')).toBeLessThan(events.findIndex((e) => e.type === 'result'));
+    } finally {
+      if (prevHome === undefined) delete process.env.CODEX_HOME;
+      else process.env.CODEX_HOME = prevHome;
+      fs.rmSync(codexHome, { recursive: true, force: true });
+    }
+  });
+
+  it('ends the turn immediately with the real cause when the app-server dies mid-turn', async () => {
+    const fake = createFakeCodexRuntime();
+    const provider = new CodexProvider({}, fake.runtime);
+    const query = provider.query({ prompt: 'prompt', cwd: '/workspace/agent' });
+    const events: ProviderEvent[] = [];
+
+    const collect = collectEvents(query.events, events);
+    await waitFor(() => fake.startCalls.length === 1);
+
+    // No pending request exists mid-turn (turn/start already resolved), so
+    // only the exitHandlers seam can end the turn — without it this parks
+    // on the waker until the 10-minute turn timeout.
+    fake.crashServer(new Error('Codex app-server exited: code=1 signal=null'));
+
+    // The generator yields the error event, then rethrows to its consumer.
+    await collect.catch(() => {});
+
+    const errors = events.filter((event) => event.type === 'error');
+    expect(errors).toHaveLength(1);
+    expect((errors[0] as { message: string }).message).toContain('app-server exited');
+  });
+});
+
+function createFakeCodexRuntime(opts: { rejectSteer?: boolean } = {}) {
+  const server = fakeServer();
+  const startCalls: TurnParams[] = [];
+  const steerCalls: Array<{ threadId: string; turnId: string; inputText: string }> = [];
+  const interruptCalls: Array<{ threadId: string; turnId: string }> = [];
+  let killed = false;
+
+  const notify = (method: string, params?: Record<string, unknown>): void => {
+    const notification: JsonRpcNotification = { method, params };
+    for (const handler of [...server.notificationHandlers]) handler(notification);
+  };
+
+  const runtime: CodexRuntimeDeps = {
+    writeCodexConfigToml: () => {},
+    spawnCodexAppServer: () => server,
+    attachCodexAutoApproval: () => {},
+    initializeCodexAppServer: async () => {},
+    startOrResumeCodexThread: async (_server, threadId) => threadId ?? 'thread-1',
+    startCodexTurn: async (_server, params) => {
+      startCalls.push(params);
+      const turnId = `turn-${startCalls.length}`;
+      notify('turn/started', { turn: { id: turnId } });
+      return turnId;
+    },
+    steerCodexTurn: async (_server, threadId, turnId, inputText) => {
+      steerCalls.push({ threadId, turnId, inputText });
+      if (opts.rejectSteer) throw new Error('steer rejected');
+    },
+    interruptCodexTurn: async (_server, threadId, turnId) => {
+      interruptCalls.push({ threadId, turnId });
+    },
+    killCodexAppServer: () => {
+      killed = true;
+    },
+  };
+
+  return {
+    runtime,
+    startCalls,
+    steerCalls,
+    interruptCalls,
+    get killed() {
+      return killed;
+    },
+    completeTurn(text: string) {
+      notify('turn/completed', { turn: { items: [{ type: 'agentMessage', text }] } });
+    },
+    crashServer(err: Error) {
+      for (const h of [...server.exitHandlers]) h(err);
+    },
+  };
+}
+
+function fakeServer(): AppServer {
+  return {
+    process: { stdin: { write: () => true }, kill: () => true },
+    readline: { close: () => {} },
+    pending: new Map(),
+    notificationHandlers: [],
+    exitHandlers: [],
+    serverRequestHandlers: [],
+  } as unknown as AppServer;
+}
+
+async function collectEvents(events: AsyncIterable<ProviderEvent>, sink: ProviderEvent[]): Promise<void> {
+  for await (const event of events) {
+    sink.push(event);
+  }
+}
+
+async function waitFor(condition: () => boolean, timeoutMs = 1000): Promise<void> {
+  const start = Date.now();
+  while (!condition()) {
+    if (Date.now() - start > timeoutMs) throw new Error('waitFor timeout');
+    await sleep(10);
+  }
+}
+
+function sleep(ms: number): Promise<void> {
+  return new Promise((resolve) => setTimeout(resolve, ms));
+}

container/agent-runner/src/providers/exchange-archive.test.ts

@@ -0,0 +1,136 @@
+import { afterEach, describe, expect, it } from 'bun:test';
+import fs from 'fs';
+import os from 'os';
+import path from 'path';
+
+import { archiveProviderExchange } from './exchange-archive.js';
+
+let tmpDir: string | null = null;
+
+afterEach(() => {
+  if (tmpDir) {
+    fs.rmSync(tmpDir, { recursive: true, force: true });
+    tmpDir = null;
+  }
+});
+
+function makeTmpDir(): string {
+  tmpDir = fs.mkdtempSync(path.join(os.tmpdir(), 'nanoclaw-archive-'));
+  return tmpDir;
+}
+
+describe('provider exchange archive', () => {
+  it('appends same-thread exchanges into one file with a single header', () => {
+    const conversationsDir = makeTmpDir();
+    const timestamp = new Date('2026-06-03T12:34:56.789Z');
+
+    const first = archiveProviderExchange({
+      conversationsDir,
+      provider: 'codex',
+      prompt: 'hello',
+      result: 'world',
+      continuation: 'thread-123',
+      status: 'completed',
+      timestamp,
+    });
+    const second = archiveProviderExchange({
+      conversationsDir,
+      provider: 'codex',
+      prompt: 'hello again',
+      result: 'world again',
+      continuation: 'thread-123',
+      status: 'completed',
+      timestamp,
+    });
+
+    // Same thread → same date-prefixed, thread-stable file, not one per exchange.
+    expect(first).toBe('2026-06-03-codex-thread-123.md');
+    expect(second).toBe(first);
+    expect(fs.readdirSync(conversationsDir)).toHaveLength(1);
+
+    const content = fs.readFileSync(path.join(conversationsDir, first!), 'utf-8');
+    // Header (thread-level metadata) written exactly once.
+    expect(content.match(/# Codex Conversation/g)).toHaveLength(1);
+    expect(content).toContain('Provider: codex');
+    expect(content).toContain('Continuation/thread id: thread-123');
+    // Both exchanges present, each with its own status line.
+    expect(content).toContain('**User**: hello');
+    expect(content).toContain('**Assistant**: world');
+    expect(content).toContain('**User**: hello again');
+    expect(content).toContain('**Assistant**: world again');
+    expect(content.match(/Status: completed/g)).toHaveLength(2);
+  });
+
+  it('writes a separate file per thread', () => {
+    const conversationsDir = makeTmpDir();
+    const timestamp = new Date('2026-06-03T12:34:56.789Z');
+
+    const a = archiveProviderExchange({
+      conversationsDir,
+      provider: 'codex',
+      prompt: 'p',
+      result: 'r',
+      continuation: 'thread-a',
+      status: 'completed',
+      timestamp,
+    });
+    const b = archiveProviderExchange({
+      conversationsDir,
+      provider: 'codex',
+      prompt: 'p',
+      result: 'r',
+      continuation: 'thread-b',
+      status: 'completed',
+      timestamp,
+    });
+
+    expect(a).toBe('2026-06-03-codex-thread-a.md');
+    expect(b).toBe('2026-06-03-codex-thread-b.md');
+    expect(fs.readdirSync(conversationsDir)).toHaveLength(2);
+  });
+
+  it('keeps the creation-date prefix stable when later exchanges land on another day', () => {
+    const conversationsDir = makeTmpDir();
+
+    const first = archiveProviderExchange({
+      conversationsDir,
+      provider: 'codex',
+      prompt: 'a',
+      result: 'b',
+      continuation: 'thread-x',
+      status: 'completed',
+      timestamp: new Date('2026-06-03T10:00:00.000Z'),
+    });
+    // A later exchange on a different day must append to the same file, not
+    // mint a new 2026-06-05-* one (the bug a naive date-from-timestamp scheme
+    // would introduce).
+    const second = archiveProviderExchange({
+      conversationsDir,
+      provider: 'codex',
+      prompt: 'c',
+      result: 'd',
+      continuation: 'thread-x',
+      status: 'completed',
+      timestamp: new Date('2026-06-05T10:00:00.000Z'),
+    });
+
+    expect(first).toBe('2026-06-03-codex-thread-x.md');
+    expect(second).toBe(first);
+    expect(fs.readdirSync(conversationsDir)).toHaveLength(1);
+  });
+
+  it('skips empty result text', () => {
+    const conversationsDir = makeTmpDir();
+    const filename = archiveProviderExchange({
+      conversationsDir,
+      provider: 'codex',
+      prompt: 'hello',
+      result: '   ',
+      continuation: 'thread-123',
+      status: 'completed',
+    });
+
+    expect(filename).toBeNull();
+    expect(fs.readdirSync(conversationsDir)).toHaveLength(0);
+  });
+});

container/agent-runner/src/providers/exchange-archive.ts

@@ -0,0 +1,105 @@
+import fs from 'fs';
+import path from 'path';
+
+/**
+ * Per-thread conversation archive for providers with no on-disk transcript —
+ * payload code, shipped with the provider that needs it. The provider's
+ * `onExchangeComplete` hook (see types.ts) calls this with each completed
+ * exchange; the runner never archives on a provider's behalf.
+ *
+ * One file per thread (keyed on the continuation id), named
+ * `<date>-<provider>-<thread>.md` and appended to as exchanges complete —
+ * mirroring the Claude path's one-file-per-session granularity and its
+ * date-prefixed, name-sortable filenames, since the Codex app-server keeps
+ * history server-side with no transcript to roll up at a compaction boundary.
+ * The date is the thread's creation day and stays stable across later appends.
+ */
+
+const DEFAULT_CONVERSATIONS_DIR = '/workspace/agent/conversations';
+
+export interface ProviderExchangeArchiveOptions {
+  provider: string;
+  prompt: string;
+  result: string | null | undefined;
+  continuation?: string;
+  status: string;
+  timestamp?: Date;
+  conversationsDir?: string;
+}
+
+/**
+ * Append a single prompt/result exchange to its thread's conversation file,
+ * writing the thread-level header once when the file is first created. Returns
+ * the (thread-stable) filename, or null when there is nothing to archive
+ * (empty result).
+ */
+export function archiveProviderExchange(options: ProviderExchangeArchiveOptions): string | null {
+  const result = options.result?.trim();
+  if (!result) return null;
+
+  const timestamp = options.timestamp ?? new Date();
+  const conversationsDir =
+    options.conversationsDir || process.env.NANOCLAW_CONVERSATIONS_DIR || DEFAULT_CONVERSATIONS_DIR;
+  fs.mkdirSync(conversationsDir, { recursive: true });
+
+  const filename = threadArchiveFilename(conversationsDir, options.provider, options.continuation, timestamp);
+  const filePath = path.join(conversationsDir, filename);
+
+  // Thread-level metadata (provider, thread id) belongs in the header, written
+  // once. Per-exchange metadata (timestamp, status) rides in each appended
+  // block. Each block leads with a blank line + `---` so the separator renders
+  // as a thematic break, not a setext heading underline on the prior line.
+  const parts: string[] = [];
+  if (!fs.existsSync(filePath)) {
+    parts.push(
+      `# ${titleCase(options.provider)} Conversation`,
+      '',
+      `Provider: ${options.provider}`,
+      `Continuation/thread id: ${options.continuation || '(none)'}`,
+    );
+  }
+  parts.push(
+    '',
+    '---',
+    '',
+    `Archived: ${timestamp.toISOString()} · Status: ${options.status}`,
+    '',
+    `**User**: ${truncate(options.prompt)}`,
+    '',
+    `**Assistant**: ${truncate(result)}`,
+    '',
+  );
+  fs.appendFileSync(filePath, parts.join('\n'));
+  return filename;
+}
+
+function threadArchiveFilename(
+  dir: string,
+  provider: string,
+  continuation: string | undefined,
+  timestamp: Date,
+): string {
+  const thread = sanitizeSlug(continuation || 'no-thread').slice(0, 48) || 'no-thread';
+  const suffix = `${sanitizeSlug(provider)}-${thread}.md`;
+  // Reuse this thread's existing file whatever day it was created; only stamp a
+  // new date when none exists. Match on the suffix after the date prefix.
+  const dated = /^\d{4}-\d{2}-\d{2}-/;
+  const existing = fs.readdirSync(dir).find((f) => dated.test(f) && f.replace(dated, '') === suffix);
+  if (existing) return existing;
+  return `${timestamp.toISOString().split('T')[0]}-${suffix}`;
+}
+
+function sanitizeSlug(value: string): string {
+  return value
+    .toLowerCase()
+    .replace(/[^a-z0-9]+/g, '-')
+    .replace(/^-+|-+$/g, '');
+}
+
+function titleCase(value: string): string {
+  return value ? value[0].toUpperCase() + value.slice(1) : 'Provider';
+}
+
+function truncate(value: string): string {
+  return value.length > 2000 ? value.slice(0, 2000) + '...' : value;
+}

container/agent-runner/src/providers/factory.test.ts

@@ -2,6 +2,7 @@ import { describe, it, expect } from 'bun:test';
 
 import { createProvider, type ProviderName } from './factory.js';
 import { ClaudeProvider } from './claude.js';
+import { CodexProvider } from './codex.js';
 import { MockProvider } from './mock.js';
 
 describe('createProvider', () => {
@@ -9,6 +10,10 @@ describe('createProvider', () => {
     expect(createProvider('claude')).toBeInstanceOf(ClaudeProvider);
   });
 
+  it('returns CodexProvider for codex', () => {
+    expect(createProvider('codex')).toBeInstanceOf(CodexProvider);
+  });
+
   it('returns MockProvider for mock', () => {
     expect(createProvider('mock')).toBeInstanceOf(MockProvider);
   });

container/agent-runner/src/providers/index.ts

@@ -3,4 +3,6 @@
 // level. Skills add a new provider by appending one import line below.
 
 import './claude.js';
+import './codex.js';
 import './mock.js';
+import './opencode.js';

container/agent-runner/src/providers/mcp-to-opencode.test.ts

@@ -0,0 +1,59 @@
+import { describe, it, expect } from 'bun:test';
+
+import { mcpServersToOpenCodeConfig } from './mcp-to-opencode.js';
+
+describe('mcpServersToOpenCodeConfig', () => {
+  it('maps nanoclaw + extra server like v2 index.ts merge', () => {
+    const servers = {
+      nanoclaw: {
+        command: 'node',
+        args: ['/app/src/mcp-tools/index.js'],
+        env: {
+          SESSION_INBOUND_DB_PATH: '/workspace/inbound.db',
+          SESSION_OUTBOUND_DB_PATH: '/workspace/outbound.db',
+          SESSION_HEARTBEAT_PATH: '/workspace/.heartbeat',
+        },
+      },
+      extra: {
+        command: 'npx',
+        args: ['-y', 'some-mcp'],
+        env: { FOO: 'bar' },
+      },
+    };
+
+    const mcp = mcpServersToOpenCodeConfig(servers);
+
+    expect(mcp.nanoclaw).toEqual({
+      type: 'local',
+      command: ['node', '/app/src/mcp-tools/index.js'],
+      environment: {
+        SESSION_INBOUND_DB_PATH: '/workspace/inbound.db',
+        SESSION_OUTBOUND_DB_PATH: '/workspace/outbound.db',
+        SESSION_HEARTBEAT_PATH: '/workspace/.heartbeat',
+      },
+      enabled: true,
+    });
+
+    expect(mcp.extra).toEqual({
+      type: 'local',
+      command: ['npx', '-y', 'some-mcp'],
+      environment: { FOO: 'bar' },
+      enabled: true,
+    });
+  });
+
+  it('omits environment when env is empty', () => {
+    const mcp = mcpServersToOpenCodeConfig({
+      x: { command: 'true', args: [], env: {} },
+    });
+    expect(mcp.x).toEqual({
+      type: 'local',
+      command: ['true'],
+      enabled: true,
+    });
+  });
+
+  it('returns empty record for undefined', () => {
+    expect(mcpServersToOpenCodeConfig(undefined)).toEqual({});
+  });
+});

container/agent-runner/src/providers/mcp-to-opencode.ts

@@ -0,0 +1,39 @@
+import type { McpServerConfig } from './types.js';
+
+/** OpenCode `mcp` entry shape (local stdio server). */
+export type OpenCodeMcpLocal = {
+  type: 'local';
+  command: string[];
+  environment?: Record<string, string>;
+  enabled: true;
+};
+
+/** OpenCode `mcp` entry shape (remote HTTP server). */
+export type OpenCodeMcpRemote = {
+  type: 'remote';
+  url: string;
+  headers?: Record<string, string>;
+  enabled: true;
+};
+
+export type OpenCodeMcpEntry = OpenCodeMcpLocal | OpenCodeMcpRemote;
+
+/**
+ * Map NanoClaw v2 MCP definitions (same shape as Claude Agent SDK) into
+ * OpenCode config `mcp` field. Stdio-only until `McpServerConfig` gains remote.
+ */
+export function mcpServersToOpenCodeConfig(
+  servers: Record<string, McpServerConfig> | undefined,
+): Record<string, OpenCodeMcpEntry> {
+  const out: Record<string, OpenCodeMcpEntry> = {};
+  if (!servers) return out;
+  for (const [name, cfg] of Object.entries(servers)) {
+    out[name] = {
+      type: 'local',
+      command: [cfg.command, ...cfg.args],
+      ...(Object.keys(cfg.env).length > 0 ? { environment: cfg.env } : {}),
+      enabled: true,
+    };
+  }
+  return out;
+}

container/agent-runner/src/providers/opencode-registration.test.ts

@@ -0,0 +1,22 @@
+/**
+ * Integration test for the opencode provider's CONTAINER-side reach-in: the self-registration
+ * import in container/agent-runner/src/providers/index.ts. Importing the barrel runs
+ * opencode.ts's top-level registerProvider('opencode', …); without that import line
+ * createProvider('opencode') throws 'Unknown provider' at runtime.
+ *
+ * Behavior, not structural, and BARREL-ONLY: it imports the real barrel (./index.js),
+ * never ./opencode.js directly, then asserts listProviderNames() contains the provider. The
+ * existing opencode.factory.test.ts imports ./opencode.js directly, so it self-registers and
+ * stays GREEN when the barrel line is deleted — a unit test, not a registration guard.
+ * This goes red if the barrel import is deleted/drifts or the barrel fails to evaluate, or if @opencode-ai/sdk is not installed (the unmocked barrel import throws) — so it also implicitly guards that dependency.
+ */
+import { describe, it, expect } from 'bun:test';
+
+import { listProviderNames } from './provider-registry.js';
+import './index.js'; // the real container provider barrel — triggers each provider's registerProvider()
+
+describe('opencode provider registration', () => {
+  it('registers opencode via the provider barrel', () => {
+    expect(listProviderNames()).toContain('opencode');
+  });
+});

container/agent-runner/src/providers/opencode.factory.test.ts

@@ -0,0 +1,10 @@
+import { describe, it, expect } from 'bun:test';
+
+import { createProvider } from './factory.js';
+import { OpenCodeProvider } from './opencode.js';
+
+describe('createProvider (opencode)', () => {
+  it('returns OpenCodeProvider for opencode', () => {
+    expect(createProvider('opencode')).toBeInstanceOf(OpenCodeProvider);
+  });
+});

container/agent-runner/src/providers/opencode.ts

@@ -0,0 +1,423 @@
+import { spawn, type ChildProcess } from 'child_process';
+
+import { createOpencodeClient, type OpencodeClient } from '@opencode-ai/sdk';
+
+import { registerProvider } from './provider-registry.js';
+import type { AgentProvider, AgentQuery, ProviderEvent, ProviderOptions, QueryInput } from './types.js';
+import { mcpServersToOpenCodeConfig } from './mcp-to-opencode.js';
+
+function log(msg: string): void {
+  console.error(`[opencode-provider] ${msg}`);
+}
+
+const SESSION_STATUS_RETRY_ERROR_AFTER = 3;
+
+/** Stale / dead OpenCode session heuristics (complement Claude-centric host patterns). */
+const STALE_SESSION_RE =
+  /no conversation found|ENOENT.*\.jsonl|session.*not found|NotFoundError|connection reset|ECONNRESET|404|event timeout/i;
+
+function killProcessTree(proc: ChildProcess): void {
+  if (!proc.pid) return;
+  try {
+    process.kill(-proc.pid, 'SIGKILL');
+  } catch {
+    try {
+      proc.kill('SIGKILL');
+    } catch {
+      /* ignore */
+    }
+  }
+}
+
+function spawnOpencodeServer(config: Record<string, unknown>, timeoutMs = 10_000): Promise<{ url: string; proc: ChildProcess }> {
+  return new Promise((resolve, reject) => {
+    const hostname = '127.0.0.1';
+    const port = 4096;
+    const proc = spawn('opencode', ['serve', `--hostname=${hostname}`, `--port=${port}`], {
+      env: {
+        ...process.env,
+        OPENCODE_CONFIG_CONTENT: JSON.stringify(config),
+      },
+      detached: true,
+    });
+
+    const id = setTimeout(() => {
+      killProcessTree(proc);
+      reject(new Error(`Timeout waiting for OpenCode server to start after ${timeoutMs}ms`));
+    }, timeoutMs);
+
+    let output = '';
+    proc.stdout?.on('data', (chunk: Buffer) => {
+      output += chunk.toString();
+      for (const line of output.split('\n')) {
+        if (line.startsWith('opencode server listening')) {
+          const match = line.match(/on\s+(https?:\/\/[^\s]+)/);
+          if (match) {
+            clearTimeout(id);
+            resolve({ url: match[1], proc });
+          }
+        }
+      }
+    });
+    proc.stderr?.on('data', (chunk: Buffer) => {
+      output += chunk.toString();
+    });
+    proc.on('exit', (code) => {
+      clearTimeout(id);
+      let msg = `OpenCode server exited with code ${code}`;
+      if (output.trim()) msg += `\nServer output: ${output}`;
+      reject(new Error(msg));
+    });
+    proc.on('error', (err) => {
+      clearTimeout(id);
+      reject(err);
+    });
+  });
+}
+
+function wrapPromptWithContext(text: string, systemInstructions?: string): string {
+  let out = text;
+  if (systemInstructions) {
+    out = `<system>\n${systemInstructions}\n</system>\n\n${out}`;
+  }
+  return out;
+}
+
+function buildOpenCodeConfig(options: ProviderOptions): Record<string, unknown> {
+  const provider = process.env.OPENCODE_PROVIDER || 'anthropic';
+  const model = process.env.OPENCODE_MODEL;
+  const smallModel = process.env.OPENCODE_SMALL_MODEL;
+  const proxyUrl = process.env.ANTHROPIC_BASE_URL;
+
+  const providerModelId = model ? model.replace(new RegExp(`^${provider}/`), '') : undefined;
+  const providerSmallModelId = smallModel ? smallModel.replace(new RegExp(`^${provider}/`), '') : undefined;
+  const modelsToRegister = [providerModelId, providerSmallModelId]
+    .filter(Boolean)
+    .filter((mid, i, a) => a.indexOf(mid as string) === i);
+
+  const providerOptions: Record<string, unknown> =
+    provider === 'anthropic'
+      ? {}
+      : {
+          [provider]: {
+            options: { apiKey: 'placeholder', baseURL: proxyUrl },
+            ...(modelsToRegister.length > 0
+              ? {
+                  models: Object.fromEntries(
+                    modelsToRegister.map((mid) => [mid, { id: mid, name: mid, tool_call: true }]),
+                  ),
+                }
+              : {}),
+          },
+        };
+
+  const mcp = mcpServersToOpenCodeConfig(options.mcpServers);
+
+  // Load shared base + per-group fragments + per-group memory through OpenCode's
+  // native instructions pipeline (session/instruction.ts). Absolute paths with
+  // globs are supported. Files are read raw — `@./...` includes are NOT expanded
+  // by OpenCode, so point at the concrete files, not at composed CLAUDE.md.
+  const instructions = [
+    '/app/CLAUDE.md',
+    '/workspace/agent/.claude-fragments/*.md',
+    '/workspace/agent/CLAUDE.local.md',
+  ];
+
+  return {
+    ...(model ? { model } : {}),
+    ...(smallModel ? { small_model: smallModel } : {}),
+    enabled_providers: [provider],
+    permission: 'allow',
+    autoupdate: false,
+    snapshot: false,
+    provider: providerOptions,
+    instructions,
+    mcp,
+  };
+}
+
+type SharedRuntime = {
+  proc: ChildProcess;
+  client: OpencodeClient;
+  stream: AsyncGenerator<{ type: string; properties: Record<string, unknown> }, void, void>;
+  streamRelease: () => void;
+};
+
+let sharedRuntime: SharedRuntime | null = null;
+let sharedConfigKey: string | null = null;
+let sharedInit: Promise<SharedRuntime> | null = null;
+
+function runtimeConfigKey(options: ProviderOptions): string {
+  return JSON.stringify({
+    mcp: mcpServersToOpenCodeConfig(options.mcpServers),
+    model: process.env.OPENCODE_MODEL,
+    small: process.env.OPENCODE_SMALL_MODEL,
+    op: process.env.OPENCODE_PROVIDER,
+  });
+}
+
+async function ensureSharedRuntime(options: ProviderOptions): Promise<SharedRuntime> {
+  const key = runtimeConfigKey(options);
+  if (sharedRuntime && sharedConfigKey === key) return sharedRuntime;
+
+  if (sharedInit) return sharedInit;
+
+  sharedInit = (async () => {
+    if (sharedRuntime) {
+      destroySharedRuntime();
+    }
+    const config = buildOpenCodeConfig(options);
+    const { url, proc } = await spawnOpencodeServer(config);
+    const client = createOpencodeClient({ baseUrl: url });
+    const sub = await client.event.subscribe();
+    const stream = sub.stream as AsyncGenerator<{ type: string; properties: Record<string, unknown> }, void, void>;
+    sharedRuntime = {
+      proc,
+      client,
+      stream,
+      streamRelease: () => {
+        void stream.return?.(undefined);
+      },
+    };
+    sharedConfigKey = key;
+    sharedInit = null;
+    return sharedRuntime;
+  })();
+
+  return sharedInit;
+}
+
+export function destroySharedRuntime(): void {
+  if (sharedRuntime) {
+    try {
+      sharedRuntime.streamRelease();
+    } catch {
+      /* ignore */
+    }
+    killProcessTree(sharedRuntime.proc);
+    sharedRuntime = null;
+    sharedConfigKey = null;
+  }
+  sharedInit = null;
+}
+
+function sessionErrorMessage(props: { error?: unknown }): string {
+  const err = props.error as { data?: { message?: string } } | undefined;
+  if (err && typeof err === 'object' && err.data && typeof err.data.message === 'string') {
+    return err.data.message;
+  }
+  return JSON.stringify(props.error) || 'OpenCode session error';
+}
+
+export class OpenCodeProvider implements AgentProvider {
+  readonly supportsNativeSlashCommands = false;
+
+  private readonly options: ProviderOptions;
+  private activeSessionId: string | undefined;
+
+  constructor(options: ProviderOptions = {}) {
+    this.options = options;
+  }
+
+  isSessionInvalid(err: unknown): boolean {
+    const msg = err instanceof Error ? err.message : String(err);
+    return STALE_SESSION_RE.test(msg);
+  }
+
+  query(input: QueryInput): AgentQuery {
+    if (input.continuation) {
+      this.activeSessionId = input.continuation;
+    } else {
+      this.activeSessionId = undefined;
+    }
+
+    const pending: string[] = [];
+    let waiting: (() => void) | null = null;
+    let ended = false;
+    let aborted = false;
+
+    const systemInstructions = input.systemContext?.instructions;
+    pending.push(wrapPromptWithContext(input.prompt, systemInstructions));
+
+    const kick = (): void => {
+      waiting?.();
+    };
+
+    const self = this;
+    const IDLE_TIMEOUT_MS = Number(process.env.OPENCODE_IDLE_TIMEOUT_MS) || 300_000;
+
+    async function* gen(): AsyncGenerator<ProviderEvent> {
+      let initYielded = false;
+      const rt = await ensureSharedRuntime(self.options);
+      const { client, stream } = rt;
+
+      while (!aborted) {
+        while (pending.length === 0 && !ended && !aborted) {
+          await new Promise<void>((resolve) => {
+            waiting = resolve;
+          });
+          waiting = null;
+        }
+
+        if (aborted) return;
+        if (pending.length === 0 && ended) return;
+
+        const text = pending.shift()!;
+        let sessionId = self.activeSessionId;
+
+        if (!sessionId) {
+          const created = await client.session.create();
+          if (created.error) {
+            throw new Error(`OpenCode: failed to create session: ${JSON.stringify(created.error)}`);
+          }
+          sessionId = created.data?.id;
+          if (!sessionId) throw new Error('OpenCode: failed to create session (no id)');
+          self.activeSessionId = sessionId;
+        }
+
+        if (!initYielded) {
+          yield { type: 'init', continuation: sessionId };
+          initYielded = true;
+        }
+
+        const promptRes = await client.session.promptAsync({
+          path: { id: sessionId },
+          body: { parts: [{ type: 'text', text }] },
+        });
+        if (promptRes.error) {
+          self.activeSessionId = undefined;
+          throw new Error(`OpenCode promptAsync: ${JSON.stringify(promptRes.error)}`);
+        }
+
+        const partTextByMessageId = new Map<string, string>();
+        const roleByMessageId = new Map<string, string>();
+        let lastEventAt = Date.now();
+        let eventTimedOut = false;
+        const timeoutCheck = setInterval(() => {
+          if (Date.now() - lastEventAt > IDLE_TIMEOUT_MS) {
+            log(`OpenCode event timeout (${IDLE_TIMEOUT_MS}ms) — clearing session ${sessionId}`);
+            eventTimedOut = true;
+            self.activeSessionId = undefined;
+            destroySharedRuntime();
+            kick();
+          }
+        }, 5000);
+
+        try {
+          turn: while (true) {
+            if (aborted) return;
+            if (eventTimedOut) {
+              throw new Error(`OpenCode event timeout (${IDLE_TIMEOUT_MS}ms)`);
+            }
+
+            const { value: ev, done } = await stream.next();
+            if (done) {
+              throw new Error('OpenCode SSE stream ended unexpectedly');
+            }
+
+            if (!ev?.type || ev.type === 'server.connected' || ev.type === 'server.heartbeat') continue;
+
+            lastEventAt = Date.now();
+            yield { type: 'activity' };
+
+            switch (ev.type) {
+              case 'message.updated': {
+                const info = ev.properties.info as { id?: string; role?: string } | undefined;
+                if (info?.id && info?.role) {
+                  roleByMessageId.set(info.id, info.role);
+                }
+                break;
+              }
+              case 'message.part.updated': {
+                const part = ev.properties.part as { type?: string; messageID?: string; text?: string } | undefined;
+                if (part?.type === 'text' && part.messageID && part.text) {
+                  partTextByMessageId.set(part.messageID, part.text);
+                }
+                break;
+              }
+              case 'permission.updated': {
+                const perm = ev.properties as { id?: string; sessionID?: string };
+                if (perm.sessionID === sessionId && perm.id) {
+                  try {
+                    await client.postSessionIdPermissionsPermissionId({
+                      path: { id: sessionId, permissionID: perm.id },
+                      body: { response: 'always' },
+                    });
+                  } catch (err) {
+                    log(`Failed to auto-reply permission: ${err instanceof Error ? err.message : String(err)}`);
+                  }
+                }
+                break;
+              }
+              case 'session.status': {
+                const props = ev.properties as {
+                  sessionID?: string;
+                  status?: { type?: string; attempt?: number; message?: string };
+                };
+                if (props.sessionID !== sessionId) break;
+                const st = props.status;
+                if (
+                  st?.type === 'retry' &&
+                  typeof st.attempt === 'number' &&
+                  st.attempt >= SESSION_STATUS_RETRY_ERROR_AFTER &&
+                  st.message
+                ) {
+                  self.activeSessionId = undefined;
+                  throw new Error(`OpenCode retry limit (${st.attempt}): ${st.message}`);
+                }
+                break;
+              }
+              case 'session.error': {
+                const props = ev.properties as { sessionID?: string; error?: unknown };
+                if (props.sessionID === sessionId || props.sessionID === undefined) {
+                  self.activeSessionId = undefined;
+                  throw new Error(sessionErrorMessage(props));
+                }
+                break;
+              }
+              case 'session.idle': {
+                const sid = (ev.properties as { sessionID?: string }).sessionID;
+                if (sid === sessionId) {
+                  break turn;
+                }
+                break;
+              }
+              default:
+                break;
+            }
+          }
+        } finally {
+          clearInterval(timeoutCheck);
+        }
+
+        let resultText = '';
+        for (const [msgId, role] of roleByMessageId) {
+          if (role === 'assistant') {
+            resultText = partTextByMessageId.get(msgId) ?? resultText;
+          }
+        }
+        yield { type: 'result', text: resultText || null };
+      }
+    }
+
+    return {
+      push: (message: string) => {
+        pending.push(wrapPromptWithContext(message, systemInstructions));
+        kick();
+      },
+      end: () => {
+        ended = true;
+        kick();
+      },
+      events: gen(),
+      abort: () => {
+        aborted = true;
+        this.activeSessionId = undefined;
+        kick();
+        destroySharedRuntime();
+      },
+    };
+  }
+}
+
+registerProvider('opencode', (opts) => new OpenCodeProvider(opts));

nanoclaw.sh

@@ -129,10 +129,10 @@ rm -f  "$PROGRESS_LOG"
 mkdir -p "$STEPS_DIR" "$LOGS_DIR"
 write_header
 
-# NanoClaw wordmark + subtitle — setup:auto will see NANOCLAW_BOOTSTRAPPED=1
-# and skip printing these again, so the flow stays visually continuous.
-printf '\n  %s%s\n' "$(bold 'Nano')" "$(brand_bold 'Claw')"
-printf '  %s\n\n' "$(dim 'Setting up your personal AI assistant')"
+# NanoClaw wordmark — clack's intro carries the "let's get you set up" framing,
+# so we don't print a subtitle here. setup:auto sees NANOCLAW_BOOTSTRAPPED=1 and
+# skips re-printing the wordmark, keeping the flow visually continuous.
+printf '\n  %s%s\n\n' "$(bold 'Nano')" "$(brand_bold 'Claw')"
 
 # ─── pre-flight: Homebrew on macOS ─────────────────────────────────────
 # setup/install-node.sh and setup/install-docker.sh both require `brew` on
@@ -190,7 +190,7 @@ BOOTSTRAP_START=$(date +%s)
 
 # One-line "why" that teaches a differentiator while the user waits.
 printf '%s  %s\n' "$(gray '│')" \
-  "$(dim "NanoClaw is small and runs entirely on your machine. Yours to modify.")"
+  "$(dim "Small. Runs on your machine. Yours to modify.")"
 spinner_start "$BOOTSTRAP_LABEL"
 
 # Run in the background so we can tick elapsed time. Capture exit code via
@@ -222,7 +222,7 @@ rm -f "$BOOTSTRAP_EXIT_FILE"
 BOOTSTRAP_DUR=$(( $(date +%s) - BOOTSTRAP_START ))
 
 if [ "$BOOTSTRAP_RC" -eq 0 ]; then
-  spinner_success "Basics installed" "$BOOTSTRAP_DUR"
+  spinner_success "Basics ready" "$BOOTSTRAP_DUR"
   write_bootstrap_entry success "$BOOTSTRAP_DUR" "$BOOTSTRAP_RAW"
 else
   spinner_failure "Couldn't install the basics" "$BOOTSTRAP_DUR"
@@ -245,7 +245,19 @@ fi
 # wipe it.
 export NANOCLAW_BOOTSTRAPPED=1
 
+# setup.sh may have just installed pnpm via npm into a prefix that's not on
+# our PATH (custom `npm config set prefix`, or the default prefix missing
+# from the shell's login PATH). Its PATH mutation doesn't propagate back
+# to us — so replay the same lookup here before the exec.
+if ! command -v pnpm >/dev/null 2>&1 && command -v npm >/dev/null 2>&1; then
+  NPM_PREFIX="$(npm config get prefix 2>/dev/null)"
+  if [ -n "$NPM_PREFIX" ] && [ -x "$NPM_PREFIX/bin/pnpm" ]; then
+    export PATH="$NPM_PREFIX/bin:$PATH"
+  fi
+fi
+
 # --silent suppresses pnpm's `> nanoclaw@2.0.0 setup:auto / > tsx setup/auto.ts`
 # preamble so the flow continues visually from "Basics installed" straight
 # into setup:auto's spinner. exec so signals (Ctrl-C) propagate directly.
-exec pnpm --silent run setup:auto
+# `-- "$@"` forwards any flags (e.g. --onecli-api-host) to setup:auto.
+exec pnpm --silent run setup:auto -- "$@"

package.json

@@ -1,6 +1,6 @@
 {
   "name": "nanoclaw",
-  "version": "2.0.4",
+  "version": "2.0.14",
   "description": "Personal Claude assistant. Lightweight, secure, customizable.",
   "type": "module",
   "packageManager": "pnpm@10.33.0",

repo-tokens/badge.svg

@@ -1,5 +1,5 @@
-<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="90" height="20" role="img" aria-label="128k tokens, 64% of context window">
-  <title>128k tokens, 64% of context window</title>
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="90" height="20" role="img" aria-label="133k tokens, 66% of context window">
+  <title>133k tokens, 66% of context window</title>
   <linearGradient id="s" x2="0" y2="100%">
     <stop offset="0" stop-color="#bbb" stop-opacity=".1"/>
     <stop offset="1" stop-opacity=".1"/>
@@ -15,8 +15,8 @@
       <g fill="#fff" text-anchor="middle" font-family="Verdana,Geneva,DejaVu Sans,sans-serif" font-size="11">
         <text aria-hidden="true" x="26" y="15" fill="#010101" fill-opacity=".3">tokens</text>
         <text x="26" y="14">tokens</text>
-        <text aria-hidden="true" x="71" y="15" fill="#010101" fill-opacity=".3">128k</text>
-        <text x="71" y="14">128k</text>
+        <text aria-hidden="true" x="71" y="15" fill="#010101" fill-opacity=".3">133k</text>
+        <text x="71" y="14">133k</text>
       </g>
     </g>
   </a>

scripts/init-first-agent.ts

@@ -48,6 +48,7 @@ import { addMember } from '../src/modules/permissions/db/agent-group-members.js'
 import { getUserRoles, grantRole } from '../src/modules/permissions/db/user-roles.js';
 import { upsertUser } from '../src/modules/permissions/db/users.js';
 import { initGroupFilesystem } from '../src/group-init.js';
+import { namespacedPlatformId } from '../src/platform-id.js';
 import type { AgentGroup, MessagingGroup } from '../src/types.js';
 
 type Role = 'owner' | 'admin' | 'member';
@@ -137,16 +138,6 @@ function namespacedUserId(channel: string, raw: string): string {
   return raw.includes(':') ? raw : `${channel}:${raw}`;
 }
 
-function namespacedPlatformId(channel: string, raw: string): string {
-  if (raw.startsWith(`${channel}:`)) return raw;
-  // Adapters using native JID format (WhatsApp: <phone>@s.whatsapp.net,
-  // <groupId>@g.us) store platform_id without a channel prefix. The '@' is
-  // the discriminator — telegram/discord platform_ids don't contain it
-  // except after a channel prefix, which is already handled above.
-  if (raw.includes('@')) return raw;
-  return `${channel}:${raw}`;
-}
-
 function generateId(prefix: string): string {
   return `${prefix}-${Date.now()}-${Math.random().toString(36).slice(2, 8)}`;
 }

setup.sh

@@ -120,6 +120,20 @@ install_deps() {
       || true
   fi
 
+  # `npm install -g` writes to npm's global prefix, which isn't always on the
+  # shell PATH — common on macOS where the user has `npm config set prefix
+  # ~/.npm-global` to avoid sudo, or on Linux where /usr/local/bin isn't in
+  # PATH. Discover the prefix and prepend its bin dir so `command -v pnpm`
+  # sees the new install.
+  if ! command -v pnpm >/dev/null 2>&1 && command -v npm >/dev/null 2>&1; then
+    local npm_prefix
+    npm_prefix=$(npm config get prefix 2>/dev/null)
+    if [ -n "$npm_prefix" ] && [ -x "$npm_prefix/bin/pnpm" ]; then
+      export PATH="$npm_prefix/bin:$PATH"
+      log "Prepended npm prefix bin to PATH: $npm_prefix/bin"
+    fi
+  fi
+
   if ! command -v pnpm >/dev/null 2>&1; then
     log "pnpm not on PATH after corepack + npm fallback"
     return

setup/add-imessage.sh

@@ -0,0 +1,160 @@
+#!/usr/bin/env bash
+#
+# Install the iMessage adapter, persist mode/creds to .env + data/env/env,
+# and restart the service. Non-interactive — the Full Disk Access walkthrough
+# (local mode) and Photon URL/key prompts (remote mode) live in
+# setup/channels/imessage.ts. Creds come in via env vars:
+#   IMESSAGE_LOCAL   'true' | 'false'  (required)
+#   IMESSAGE_ENABLED 'true'            (required when IMESSAGE_LOCAL=true)
+#   IMESSAGE_SERVER_URL                (required when IMESSAGE_LOCAL=false)
+#   IMESSAGE_API_KEY                   (required when IMESSAGE_LOCAL=false)
+#
+# Emits exactly one status block on stdout (ADD_IMESSAGE) at the end.
+set -euo pipefail
+
+PROJECT_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+cd "$PROJECT_ROOT"
+
+# Keep in sync with .claude/skills/add-imessage/SKILL.md.
+ADAPTER_VERSION="chat-adapter-imessage@0.1.1"
+
+# Resolve which remote carries the channels branch — handles forks where
+# upstream lives on a different remote than `origin`.
+# shellcheck source=setup/lib/channels-remote.sh
+source "$PROJECT_ROOT/setup/lib/channels-remote.sh"
+CHANNELS_REMOTE=$(resolve_channels_remote)
+CHANNELS_BRANCH="${CHANNELS_REMOTE}/channels"
+
+emit_status() {
+  local status=$1 error=${2:-}
+  local already=${ADAPTER_ALREADY_INSTALLED:-false}
+  local mode=${IMESSAGE_LOCAL:-}
+  echo "=== NANOCLAW SETUP: ADD_IMESSAGE ==="
+  echo "STATUS: ${status}"
+  echo "ADAPTER_VERSION: ${ADAPTER_VERSION}"
+  echo "ADAPTER_ALREADY_INSTALLED: ${already}"
+  [ -n "$mode" ] && echo "MODE: $([ "$mode" = "true" ] && echo local || echo remote)"
+  [ -n "$error" ] && echo "ERROR: ${error}"
+  echo "=== END ==="
+}
+
+log() { echo "[add-imessage] $*" >&2; }
+
+# Validate creds based on mode.
+if [ -z "${IMESSAGE_LOCAL:-}" ]; then
+  emit_status failed "IMESSAGE_LOCAL env var not set (expected true|false)"
+  exit 1
+fi
+if [ "${IMESSAGE_LOCAL}" = "true" ]; then
+  if [ -z "${IMESSAGE_ENABLED:-}" ]; then
+    emit_status failed "IMESSAGE_ENABLED env var not set for local mode"
+    exit 1
+  fi
+  if [ "$(uname -s)" != "Darwin" ]; then
+    emit_status failed "local mode requires macOS"
+    exit 1
+  fi
+else
+  if [ -z "${IMESSAGE_SERVER_URL:-}" ]; then
+    emit_status failed "IMESSAGE_SERVER_URL env var not set for remote mode"
+    exit 1
+  fi
+  if [ -z "${IMESSAGE_API_KEY:-}" ]; then
+    emit_status failed "IMESSAGE_API_KEY env var not set for remote mode"
+    exit 1
+  fi
+fi
+
+need_install() {
+  [ ! -f src/channels/imessage.ts ] && return 0
+  ! grep -q "^import './imessage.js';" src/channels/index.ts 2>/dev/null && return 0
+  return 1
+}
+
+ADAPTER_ALREADY_INSTALLED=true
+if need_install; then
+  ADAPTER_ALREADY_INSTALLED=false
+  log "Fetching channels branch…"
+  git fetch "$CHANNELS_REMOTE" channels >&2 2>/dev/null || {
+    emit_status failed "git fetch ${CHANNELS_REMOTE} channels failed"
+    exit 1
+  }
+
+  log "Copying adapter from ${CHANNELS_BRANCH}…"
+  git show "${CHANNELS_BRANCH}:src/channels/imessage.ts" > src/channels/imessage.ts
+
+  # Append self-registration import if missing.
+  if ! grep -q "^import './imessage.js';" src/channels/index.ts; then
+    echo "import './imessage.js';" >> src/channels/index.ts
+  fi
+
+  log "Installing ${ADAPTER_VERSION}…"
+  pnpm install "${ADAPTER_VERSION}" >&2 2>/dev/null || {
+    emit_status failed "pnpm install ${ADAPTER_VERSION} failed"
+    exit 1
+  }
+
+  log "Building…"
+  pnpm run build >&2 2>/dev/null || {
+    emit_status failed "pnpm run build failed"
+    exit 1
+  }
+else
+  log "Adapter files already installed — skipping install phase."
+fi
+
+touch .env
+upsert_env() {
+  local key=$1 value=$2
+  if grep -q "^${key}=" .env; then
+    awk -v k="$key" -v v="$value" \
+        'BEGIN{FS=OFS="="} $1==k {print k "=" v; next} {print}' \
+      .env > .env.tmp && mv .env.tmp .env
+  else
+    echo "${key}=${value}" >> .env
+  fi
+}
+
+remove_env() {
+  local key=$1
+  if grep -q "^${key}=" .env 2>/dev/null; then
+    grep -v "^${key}=" .env > .env.tmp && mv .env.tmp .env
+  fi
+}
+
+# Write the canonical keys for the chosen mode, strip the opposite mode's
+# keys so stale values can't confuse the adapter's factory.
+upsert_env IMESSAGE_LOCAL "$IMESSAGE_LOCAL"
+if [ "$IMESSAGE_LOCAL" = "true" ]; then
+  upsert_env IMESSAGE_ENABLED "$IMESSAGE_ENABLED"
+  remove_env IMESSAGE_SERVER_URL
+  remove_env IMESSAGE_API_KEY
+else
+  upsert_env IMESSAGE_SERVER_URL "$IMESSAGE_SERVER_URL"
+  upsert_env IMESSAGE_API_KEY "$IMESSAGE_API_KEY"
+  remove_env IMESSAGE_ENABLED
+fi
+
+# Container reads from data/env/env (the host mounts it).
+mkdir -p data/env
+cp .env data/env/env
+
+log "Restarting service so the new adapter picks up the creds…"
+# shellcheck source=setup/lib/install-slug.sh
+source "$PROJECT_ROOT/setup/lib/install-slug.sh"
+case "$(uname -s)" in
+  Darwin)
+    launchctl kickstart -k "gui/$(id -u)/$(launchd_label)" >&2 2>/dev/null || true
+    ;;
+  Linux)
+    systemctl --user restart "$(systemd_unit)" >&2 2>/dev/null \
+      || sudo systemctl restart "$(systemd_unit)" >&2 2>/dev/null \
+      || true
+    ;;
+esac
+
+# Give the adapter a moment to open chat.db (local) or handshake with
+# Photon (remote) before emitting success.
+sleep 3
+
+emit_status success

setup/add-signal.sh

@@ -0,0 +1,95 @@
+#!/usr/bin/env bash
+#
+# Install the Signal adapter in an already-running NanoClaw checkout.
+# Non-interactive — the operator-facing "install signal-cli" + QR scan
+# live in setup/channels/signal.ts. This script only:
+#
+#   1. Fetches src/channels/signal.ts + signal.test.ts from the channels
+#      branch.
+#   2. Appends the self-registration import to src/channels/index.ts.
+#   3. Installs qrcode (for setup-flow QR rendering — adapter itself has
+#      no npm deps).
+#   4. Builds.
+#
+# SIGNAL_ACCOUNT is persisted separately by the driver once signal-cli
+# link has produced a number; that keeps this script idempotent and
+# re-runnable without re-auth.
+#
+# Emits exactly one status block on stdout (ADD_SIGNAL) at the end. All
+# chatty progress goes to stderr so setup:auto's raw-log capture sees
+# the full story without cluttering the final block for the parser.
+set -euo pipefail
+
+PROJECT_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+cd "$PROJECT_ROOT"
+
+# Keep in sync with .claude/skills/add-signal/SKILL.md.
+QRCODE_VERSION="qrcode@1.5.4"
+QRCODE_TYPES_VERSION="@types/qrcode@1.5.6"
+
+# shellcheck source=setup/lib/channels-remote.sh
+source "$PROJECT_ROOT/setup/lib/channels-remote.sh"
+CHANNELS_REMOTE=$(resolve_channels_remote)
+CHANNELS_BRANCH="${CHANNELS_REMOTE}/channels"
+
+emit_status() {
+  local status=$1 error=${2:-}
+  local already=${ADAPTER_ALREADY_INSTALLED:-false}
+  echo "=== NANOCLAW SETUP: ADD_SIGNAL ==="
+  echo "STATUS: ${status}"
+  echo "ADAPTER_ALREADY_INSTALLED: ${already}"
+  [ -n "$error" ] && echo "ERROR: ${error}"
+  echo "=== END ==="
+}
+
+log() { echo "[add-signal] $*" >&2; }
+
+need_install() {
+  [ ! -f src/channels/signal.ts ] && return 0
+  ! grep -q "^import './signal.js';" src/channels/index.ts 2>/dev/null && return 0
+  return 1
+}
+
+ADAPTER_ALREADY_INSTALLED=true
+if need_install; then
+  ADAPTER_ALREADY_INSTALLED=false
+  log "Fetching channels branch…"
+  git fetch "$CHANNELS_REMOTE" channels >&2 2>/dev/null || {
+    emit_status failed "git fetch ${CHANNELS_REMOTE} channels failed"
+    exit 1
+  }
+
+  log "Copying adapter files from ${CHANNELS_BRANCH}…"
+  for f in \
+    src/channels/signal.ts \
+    src/channels/signal.test.ts
+  do
+    git show "${CHANNELS_BRANCH}:$f" > "$f" || {
+      emit_status failed "git show ${CHANNELS_BRANCH}:$f failed"
+      exit 1
+    }
+  done
+
+  if ! grep -q "^import './signal.js';" src/channels/index.ts; then
+    echo "import './signal.js';" >> src/channels/index.ts
+  fi
+fi
+
+# qrcode is needed by setup/signal-auth.ts to render the linking URL as a
+# terminal QR. Install idempotently — if it's already present (e.g. from a
+# prior WhatsApp install) pnpm is a no-op.
+if ! node -e "require.resolve('qrcode')" >/dev/null 2>&1; then
+  log "Installing ${QRCODE_VERSION}…"
+  pnpm install "${QRCODE_VERSION}" "${QRCODE_TYPES_VERSION}" >&2 2>/dev/null || {
+    emit_status failed "pnpm install ${QRCODE_VERSION} failed"
+    exit 1
+  }
+fi
+
+log "Building…"
+pnpm run build >&2 2>/dev/null || {
+  emit_status failed "pnpm run build failed"
+  exit 1
+}
+
+emit_status success

setup/add-slack.sh

@@ -0,0 +1,125 @@
+#!/usr/bin/env bash
+#
+# Install the Slack adapter, persist SLACK_BOT_TOKEN + SLACK_SIGNING_SECRET to
+# .env + data/env/env, and restart the service. Non-interactive — the
+# operator-facing app creation walkthrough + credential paste live in
+# setup/channels/slack.ts. Credentials come in via env vars:
+# SLACK_BOT_TOKEN, SLACK_SIGNING_SECRET.
+#
+# Emits exactly one status block on stdout (ADD_SLACK) at the end. All chatty
+# progress messages go to stderr so setup:auto's raw-log capture sees the full
+# story without cluttering the final block for the parser.
+set -euo pipefail
+
+PROJECT_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+cd "$PROJECT_ROOT"
+
+# Keep in sync with .claude/skills/add-slack/SKILL.md.
+ADAPTER_VERSION="@chat-adapter/slack@4.26.0"
+
+# Resolve which remote carries the channels branch — handles forks where
+# upstream lives on a different remote than `origin`.
+# shellcheck source=setup/lib/channels-remote.sh
+source "$PROJECT_ROOT/setup/lib/channels-remote.sh"
+CHANNELS_REMOTE=$(resolve_channels_remote)
+CHANNELS_BRANCH="${CHANNELS_REMOTE}/channels"
+
+emit_status() {
+  local status=$1 error=${2:-}
+  local already=${ADAPTER_ALREADY_INSTALLED:-false}
+  echo "=== NANOCLAW SETUP: ADD_SLACK ==="
+  echo "STATUS: ${status}"
+  echo "ADAPTER_VERSION: ${ADAPTER_VERSION}"
+  echo "ADAPTER_ALREADY_INSTALLED: ${already}"
+  [ -n "$error" ] && echo "ERROR: ${error}"
+  echo "=== END ==="
+}
+
+log() { echo "[add-slack] $*" >&2; }
+
+if [ -z "${SLACK_BOT_TOKEN:-}" ]; then
+  emit_status failed "SLACK_BOT_TOKEN env var not set"
+  exit 1
+fi
+if [ -z "${SLACK_SIGNING_SECRET:-}" ]; then
+  emit_status failed "SLACK_SIGNING_SECRET env var not set"
+  exit 1
+fi
+
+need_install() {
+  [ ! -f src/channels/slack.ts ] && return 0
+  ! grep -q "^import './slack.js';" src/channels/index.ts 2>/dev/null && return 0
+  return 1
+}
+
+ADAPTER_ALREADY_INSTALLED=true
+if need_install; then
+  ADAPTER_ALREADY_INSTALLED=false
+  log "Fetching channels branch…"
+  git fetch "$CHANNELS_REMOTE" channels >&2 2>/dev/null || {
+    emit_status failed "git fetch ${CHANNELS_REMOTE} channels failed"
+    exit 1
+  }
+
+  log "Copying adapter from ${CHANNELS_BRANCH}…"
+  git show "${CHANNELS_BRANCH}:src/channels/slack.ts" > src/channels/slack.ts
+
+  # Append self-registration import if missing.
+  if ! grep -q "^import './slack.js';" src/channels/index.ts; then
+    echo "import './slack.js';" >> src/channels/index.ts
+  fi
+
+  log "Installing ${ADAPTER_VERSION}…"
+  pnpm install "${ADAPTER_VERSION}" >&2 2>/dev/null || {
+    emit_status failed "pnpm install ${ADAPTER_VERSION} failed"
+    exit 1
+  }
+
+  log "Building…"
+  pnpm run build >&2 2>/dev/null || {
+    emit_status failed "pnpm run build failed"
+    exit 1
+  }
+else
+  log "Adapter files already installed — skipping install phase."
+fi
+
+# Persist credentials. auto.ts validates via auth.test before this point, so
+# bad values here would be an internal bug rather than operator input.
+touch .env
+upsert_env() {
+  local key=$1 value=$2
+  if grep -q "^${key}=" .env; then
+    awk -v k="$key" -v v="$value" \
+        'BEGIN{FS=OFS="="} $1==k {print k "=" v; next} {print}' \
+      .env > .env.tmp && mv .env.tmp .env
+  else
+    echo "${key}=${value}" >> .env
+  fi
+}
+upsert_env SLACK_BOT_TOKEN "$SLACK_BOT_TOKEN"
+upsert_env SLACK_SIGNING_SECRET "$SLACK_SIGNING_SECRET"
+
+# Container reads from data/env/env (the host mounts it).
+mkdir -p data/env
+cp .env data/env/env
+
+log "Restarting service so the new adapter picks up the credentials…"
+# shellcheck source=setup/lib/install-slug.sh
+source "$PROJECT_ROOT/setup/lib/install-slug.sh"
+case "$(uname -s)" in
+  Darwin)
+    launchctl kickstart -k "gui/$(id -u)/$(launchd_label)" >&2 2>/dev/null || true
+    ;;
+  Linux)
+    systemctl --user restart "$(systemd_unit)" >&2 2>/dev/null \
+      || sudo systemctl restart "$(systemd_unit)" >&2 2>/dev/null \
+      || true
+    ;;
+esac
+
+# Give the Slack adapter a moment to finish starting the webhook listener
+# before emitting success.
+sleep 3
+
+emit_status success

setup/auto.ts

@@ -22,23 +22,33 @@
  * headless `claude -p` call for IANA-zone resolution.
  */
 import { spawn, spawnSync } from 'child_process';
+import fs from 'fs';
+import path from 'path';
 
 import * as p from '@clack/prompts';
 import k from 'kleur';
 
 import { runDiscordChannel } from './channels/discord.js';
+import { runIMessageChannel } from './channels/imessage.js';
+import { runSignalChannel } from './channels/signal.js';
+import { runSlackChannel } from './channels/slack.js';
 import { runTeamsChannel } from './channels/teams.js';
 import { runTelegramChannel } from './channels/telegram.js';
 import { runWhatsAppChannel } from './channels/whatsapp.js';
 import { pingCliAgent, type PingResult } from './lib/agent-ping.js';
 import { brightSelect } from './lib/bright-select.js';
 import { offerClaudeAssist } from './lib/claude-assist.js';
-import { runWindowedStep } from './lib/windowed-runner.js';
-import { getLaunchdLabel, getSystemdUnit } from '../src/install-slug.js';
 import {
-  claudeCliAvailable,
-  resolveTimezoneViaClaude,
-} from './lib/tz-from-claude.js';
+  applyToEnv,
+  parseFlags,
+  printHelp,
+  readFromEnv,
+} from './lib/setup-config-parse.js';
+import { runAdvancedScreen } from './lib/setup-config-screen.js';
+import { runWindowedStep } from './lib/windowed-runner.js';
+import { pollHealth } from './onecli.js';
+import { getLaunchdLabel, getSystemdUnit } from '../src/install-slug.js';
+import { claudeCliAvailable, resolveTimezoneViaClaude } from './lib/tz-from-claude.js';
 import * as setupLog from './logs.js';
 import { ensureAnswer, fail, runQuietChild, runQuietStep } from './lib/runner.js';
 import { emit as phEmit } from './lib/diagnostics.js';
@@ -48,11 +58,48 @@ import { isValidTimezone } from '../src/timezone.js';
 const CLI_AGENT_NAME = 'Terminal Agent';
 const RUN_START = Date.now();
 
+type ChannelChoice = 'telegram' | 'discord' | 'whatsapp' | 'signal' | 'teams' | 'slack' | 'imessage' | 'skip';
+
 async function main(): Promise<void> {
+  // Parse CLI flags first — `--help` short-circuits before we render anything,
+  // and flag values get folded into process.env so existing step code reading
+  // NANOCLAW_* sees them unchanged.
+  const flagResult = parseFlags(process.argv.slice(2));
+  if (flagResult.help) {
+    printHelp();
+    process.exit(0);
+  }
+  if (flagResult.errors.length > 0) {
+    for (const err of flagResult.errors) console.error(`error: ${err}`);
+    console.error('');
+    console.error('Run with --help for the full list of supported flags.');
+    process.exit(1);
+  }
+  let configValues = { ...readFromEnv(), ...flagResult.values };
+  applyToEnv(configValues);
+
   printIntro();
   initProgressionLog();
   phEmit('auto_started');
 
+  // Welcome menu — default path or open advanced overrides before any setup
+  // work begins. Default lands on standard so Enter is the happy path.
+  const startChoice = ensureAnswer(
+    await brightSelect<'default' | 'advanced'>({
+      message: 'How would you like to begin?',
+      options: [
+        { value: 'default', label: 'Standard setup' },
+        { value: 'advanced', label: 'Advanced', hint: 'override defaults' },
+      ],
+      initialValue: 'default',
+    }),
+  ) as 'default' | 'advanced';
+  setupLog.userInput('start_choice', startChoice);
+  if (startChoice === 'advanced') {
+    configValues = await runAdvancedScreen(configValues);
+    applyToEnv(configValues);
+  }
+
   const skip = new Set(
     (process.env.NANOCLAW_SKIP ?? '')
       .split(',')
@@ -75,12 +122,7 @@ async function main(): Promise<void> {
   }
 
   if (!skip.has('container')) {
-    p.log.message(
-      dimWrap(
-        'Your assistant lives in its own sandbox. It can only see what you explicitly share.',
-        4,
-      ),
-    );
+    p.log.message(dimWrap('Your assistant lives in its own sandbox. It can only see what you explicitly share.', 4));
     p.log.message(
       dimWrap(
         'The first build pulls a base image and installs a few tools. On a fresh machine this usually takes 3–10 minutes.',
@@ -125,57 +167,95 @@ async function main(): Promise<void> {
       ),
     );
 
-    // Respect an existing OneCLI install. Re-running the installer would
-    // rebind the listener and knock any other app using that gateway
-    // offline — confirm with the user before doing that.
-    const existing = detectExistingOnecli();
-    let reuse = false;
-    if (existing) {
-      const choice = ensureAnswer(
-        await brightSelect({
-          message: `Found an existing OneCLI at ${existing.apiHost}. What would you like to do?`,
-          options: [
-            {
-              value: 'reuse',
-              label: 'Use the existing instance',
-              hint: 'recommended — keeps other apps bound to this vault working',
-            },
-            {
-              value: 'fresh',
-              label: 'Install a fresh instance for NanoClaw',
-              hint: 'reinstalls onecli; other apps may need to reconnect',
-            },
-          ],
-        }),
-      ) as 'reuse' | 'fresh';
-      setupLog.userInput('onecli_choice', choice);
-      reuse = choice === 'reuse';
-    }
+    const remoteHost = process.env.NANOCLAW_ONECLI_API_HOST?.trim();
 
-    const res = await runQuietStep(
-      'onecli',
-      {
-        running: reuse
-          ? 'Hooking up to your existing OneCLI…'
-          : "Setting up OneCLI, your agent's vault…",
-        done: 'OneCLI vault ready.',
-      },
-      reuse ? ['--reuse'] : [],
-    );
-    if (!res.ok) {
-      const err = res.terminal?.fields.ERROR;
-      if (err === 'onecli_not_on_path_after_install') {
+    if (remoteHost) {
+      // Advanced-settings override: user has already named a remote vault,
+      // so skip the local-vs-fresh prompt entirely. Health-check it here
+      // rather than letting the step fail silently — a typo in the URL is a
+      // common mistake and the answer is human-fixable.
+      const s = p.spinner();
+      s.start(`Checking remote OneCLI at ${remoteHost}…`);
+      const healthy = await pollHealth(remoteHost, 5000);
+      if (!healthy) {
+        s.stop(`Couldn't reach OneCLI at ${remoteHost}.`, 1);
         await fail(
           'onecli',
-          'OneCLI was installed but your shell needs to refresh to see it.',
-          'Open a new shell or run `export PATH="$HOME/.local/bin:$PATH"`, then retry.',
+          `Couldn't reach OneCLI at ${remoteHost}.`,
+          'Check the URL and that OneCLI is running on the remote machine, then retry.',
         );
       }
-      await fail(
+      s.stop('Remote OneCLI is reachable.');
+
+      const res = await runQuietStep(
         'onecli',
-        `Couldn't set up OneCLI (${err ?? 'unknown error'}).`,
-        'Make sure curl is installed and ~/.local/bin is writable, then retry.',
+        {
+          running: `Connecting to remote OneCLI at ${remoteHost}…`,
+          done: 'OneCLI vault ready.',
+        },
+        ['--remote-url', remoteHost],
       );
+      if (!res.ok) {
+        const err = res.terminal?.fields.ERROR;
+        await fail(
+          'onecli',
+          `Couldn't connect to remote OneCLI (${err ?? 'unknown error'}).`,
+          'Check the URL and that OneCLI is running on the remote machine, then retry.',
+        );
+      }
+    } else {
+      // Respect an existing OneCLI install. Re-running the installer would
+      // rebind the listener and knock any other app using that gateway
+      // offline — confirm with the user before doing that.
+      const existing = detectExistingOnecli();
+      let reuse = false;
+      if (existing) {
+        const choice = ensureAnswer(
+          await brightSelect({
+            message: `Found an existing OneCLI at ${existing.apiHost}. What would you like to do?`,
+            options: [
+              {
+                value: 'reuse',
+                label: 'Use the existing instance',
+                hint: 'recommended — keeps other apps bound to this vault working',
+              },
+              {
+                value: 'fresh',
+                label: 'Install a fresh instance for NanoClaw',
+                hint: 'reinstalls onecli; other apps may need to reconnect',
+              },
+            ],
+          }),
+        ) as 'reuse' | 'fresh';
+        setupLog.userInput('onecli_choice', choice);
+        reuse = choice === 'reuse';
+      }
+
+      const res = await runQuietStep(
+        'onecli',
+        {
+          running: reuse
+            ? 'Hooking up to your existing OneCLI…'
+            : "Setting up OneCLI, your agent's vault…",
+          done: 'OneCLI vault ready.',
+        },
+        reuse ? ['--reuse'] : [],
+      );
+      if (!res.ok) {
+        const err = res.terminal?.fields.ERROR;
+        if (err === 'onecli_not_on_path_after_install') {
+          await fail(
+            'onecli',
+            'OneCLI was installed but your shell needs to refresh to see it.',
+            'Open a new shell or run `export PATH="$HOME/.local/bin:$PATH"`, then retry.',
+          );
+        }
+        await fail(
+          'onecli',
+          `Couldn't set up OneCLI (${err ?? 'unknown error'}).`,
+          'Make sure curl is installed and ~/.local/bin is writable, then retry.',
+        );
+      }
     }
   }
 
@@ -204,21 +284,12 @@ async function main(): Promise<void> {
       done: 'NanoClaw is running.',
     });
     if (!res.ok) {
-      await fail(
-        'service',
-        "Couldn't start NanoClaw.",
-        'See logs/nanoclaw.error.log for details.',
-      );
+      await fail('service', "Couldn't start NanoClaw.", 'See logs/nanoclaw.error.log for details.');
     }
     if (res.terminal?.fields.DOCKER_GROUP_STALE === 'true') {
-      p.log.warn(
-        "NanoClaw's permissions need a tweak before it can reach Docker.",
-      );
+      p.log.warn("NanoClaw's permissions need a tweak before it can reach Docker.");
       p.log.message(
-        k.dim(
-          '  sudo setfacl -m u:$(whoami):rw /var/run/docker.sock\n' +
-            '  systemctl --user restart nanoclaw',
-        ),
+        '  sudo setfacl -m u:$(whoami):rw /var/run/docker.sock\n' + `  systemctl --user restart ${getSystemdUnit()}`,
       );
     }
   }
@@ -248,10 +319,33 @@ async function main(): Promise<void> {
       );
     }
     if (!skip.has('first-chat')) {
+      p.log.message(
+        dimWrap(
+          "Your assistant runs in an isolated sandbox. I'm going to send it a quick test message (ping) and wait for a reply (pong) to confirm it's responding. First startup typically takes 30–60 seconds while the sandbox warms up.",
+          4,
+        ),
+      );
       const ping = await confirmAssistantResponds();
       if (ping === 'ok') {
         phEmit('first_chat_ready');
-        await runFirstChat();
+        const next = ensureAnswer(
+          await p.select({
+            message: 'What next?',
+            options: [
+              {
+                value: 'continue',
+                label: 'Continue with setup',
+                hint: 'recommended',
+              },
+              {
+                value: 'chat',
+                label: 'Pause here and chat with your agent from the terminal',
+              },
+            ],
+          }),
+        ) as 'continue' | 'chat';
+        setupLog.userInput('first_chat_choice', next);
+        if (next === 'chat') await runFirstChat();
       } else {
         phEmit('first_chat_failed', { reason: ping });
         renderPingFailureNote(ping);
@@ -260,7 +354,7 @@ async function main(): Promise<void> {
           msg:
             ping === 'socket_error'
               ? "NanoClaw service isn't listening on its CLI socket."
-              : "No reply from the assistant within 30 seconds.",
+              : 'No reply from the assistant within 30 seconds.',
           hint:
             ping === 'socket_error'
               ? 'Socket at data/cli.sock did not accept a connection.'
@@ -274,8 +368,7 @@ async function main(): Promise<void> {
     await runTimezoneStep();
   }
 
-  let channelChoice: 'telegram' | 'discord' | 'whatsapp' | 'teams' | 'skip' =
-    'skip';
+  let channelChoice: ChannelChoice = 'skip';
   if (!skip.has('channel')) {
     channelChoice = await askChannelChoice();
     if (channelChoice === 'telegram') {
@@ -284,12 +377,18 @@ async function main(): Promise<void> {
       await runDiscordChannel(displayName!);
     } else if (channelChoice === 'whatsapp') {
       await runWhatsAppChannel(displayName!);
+    } else if (channelChoice === 'signal') {
+      await runSignalChannel(displayName!);
     } else if (channelChoice === 'teams') {
       await runTeamsChannel(displayName!);
+    } else if (channelChoice === 'slack') {
+      await runSlackChannel(displayName!);
+    } else if (channelChoice === 'imessage') {
+      await runIMessageChannel(displayName!);
     } else {
       p.log.info(
         wrapForGutter(
-          'No messaging app for now. You can add one later (like Telegram, Discord, WhatsApp, Teams, or Slack).',
+          'No messaging app for now. You can add one later (like Telegram, Discord, WhatsApp, Teams, Slack, or iMessage).',
           4,
         ),
       );
@@ -305,7 +404,7 @@ async function main(): Promise<void> {
     if (!res.ok) {
       const notes: string[] = [];
       if (res.terminal?.fields.CREDENTIALS !== 'configured') {
-        notes.push('• Your Claude account isn\'t connected. Re-run setup and try again.');
+        notes.push("• Your Claude account isn't connected. Re-run setup and try again.");
       }
       const service = res.terminal?.fields.SERVICE;
       if (service === 'running_other_checkout') {
@@ -331,7 +430,9 @@ async function main(): Promise<void> {
         }
       }
       if (!res.terminal?.fields.CONFIGURED_CHANNELS) {
-        notes.push('• Want to chat from your phone? Add a messaging app with `/add-telegram`, `/add-slack`, or `/add-discord`.');
+        notes.push(
+          '• Want to chat from your phone? Add a messaging app with `/add-telegram`, `/add-slack`, or `/add-discord`.',
+        );
       }
       if (notes.length > 0) {
         p.note(notes.join('\n'), "What's left");
@@ -365,9 +466,7 @@ async function main(): Promise<void> {
     ['Open Claude Code:', 'claude'],
   ];
   const labelWidth = Math.max(...rows.map(([l]) => l.length));
-  const nextSteps = rows
-    .map(([l, c]) => `${k.cyan(l.padEnd(labelWidth))}  ${c}`)
-    .join('\n');
+  const nextSteps = rows.map(([l, c]) => `${k.cyan(l.padEnd(labelWidth))}  ${c}`).join('\n');
   p.note(nextSteps, 'Try these');
 
   // Always-on warning goes before the "check your DMs" directive so the
@@ -389,19 +488,14 @@ async function main(): Promise<void> {
     // that the welcome-message signal was too easy to miss. Use p.note so it
     // renders with a visible box, cyan-bold the directive line, and put it
     // as the last thing before outro.
-    p.note(
-      `${brandBold('→')} ${k.bold(`Check your ${dmTarget} — your assistant is saying hi.`)}`,
-      'Go say hi',
-    );
+    p.note(`${brandBold('→')} ${k.bold(`Check your ${dmTarget} — your assistant is saying hi.`)}`, 'Go say hi');
     p.outro(k.green("You're set."));
   } else {
     p.outro(k.green("You're ready! Chat with `pnpm run chat hi`."));
   }
 }
 
-function channelDmLabel(
-  choice: 'telegram' | 'discord' | 'whatsapp' | 'teams' | 'skip',
-): string | null {
+function channelDmLabel(choice: ChannelChoice): string | null {
   switch (choice) {
     case 'telegram':
       return 'Telegram';
@@ -409,8 +503,17 @@ function channelDmLabel(
       return 'Discord DMs';
     case 'whatsapp':
       return 'WhatsApp';
+    case 'signal':
+      return 'Signal';
     case 'teams':
       return 'Teams';
+    case 'imessage':
+      return 'iMessage';
+    case 'slack':
+      // Slack install doesn't wire an agent or send a welcome DM — the
+      // driver prints its own "finish in your Slack app" note. Falling
+      // through to null avoids a misleading "check your Slack DMs" banner.
+      return null;
     default:
       return null;
   }
@@ -442,13 +545,11 @@ async function confirmAssistantResponds(): Promise<PingResult> {
   const elapsed = Math.round((Date.now() - start) / 1000);
   const suffix = ` (${elapsed}s)`;
   if (result === 'ok') {
-    s.stop(`${fitToWidth('Your assistant is ready.', suffix)}${k.dim(suffix)}`);
+    s.stop(`${k.bold(fitToWidth('Your assistant is ready.', suffix))}${k.dim(suffix)}`);
   } else {
     const msg =
-      result === 'socket_error'
-        ? "Couldn't reach the NanoClaw service."
-        : "Your assistant didn't reply in time.";
-    s.stop(`${fitToWidth(msg, suffix)}${k.dim(suffix)}`, 1);
+      result === 'socket_error' ? "Couldn't reach the NanoClaw service." : "Your assistant didn't reply in time.";
+    s.stop(`${k.bold(fitToWidth(msg, suffix))}${k.dim(suffix)}`, 1);
   }
   return result;
 }
@@ -462,8 +563,8 @@ function renderPingFailureNote(result: PingResult): void {
             6,
           ),
           '',
-          k.dim(`  macOS:  launchctl kickstart -k gui/$(id -u)/${getLaunchdLabel()}`),
-          k.dim(`  Linux:  systemctl --user restart ${getSystemdUnit()}`),
+          `  macOS:  launchctl kickstart -k gui/$(id -u)/${getLaunchdLabel()}`,
+          `  Linux:  systemctl --user restart ${getSystemdUnit()}`,
         ].join('\n')
       : wrapForGutter(
           'No reply from your assistant within 30 seconds. Check `logs/nanoclaw.log` for clues, then try `pnpm run chat hi`.',
@@ -503,9 +604,7 @@ async function runFirstChat(): Promise<void> {
         message: first
           ? 'Try a quick hello — or press Enter to continue setup'
           : 'Another message? Press Enter to continue setup',
-        placeholder: first
-          ? 'e.g. "hi, what can you do?"'
-          : 'press Enter to continue',
+        placeholder: first ? 'e.g. "hi, what can you do?"' : 'press Enter to continue',
       }),
     );
     first = false;
@@ -521,11 +620,9 @@ function sendChatMessage(message: string): Promise<void> {
     // agent's reply reads as a clean block under the prompt. Splitting on
     // whitespace mirrors `pnpm run chat hello world` — chat.ts joins argv
     // with spaces on the far side.
-    const child = spawn(
-      'pnpm',
-      ['--silent', 'run', 'chat', ...message.split(/\s+/)],
-      { stdio: ['ignore', 'inherit', 'inherit'] },
-    );
+    const child = spawn('pnpm', ['--silent', 'run', 'chat', ...message.split(/\s+/)], {
+      stdio: ['ignore', 'inherit', 'inherit'],
+    });
     child.on('close', () => resolve());
     child.on('error', () => resolve());
   });
@@ -540,6 +637,16 @@ async function runAuthStep(): Promise<void> {
     return;
   }
 
+  // Custom Anthropic-compatible endpoint flow. Both URL and token must be set;
+  // OneCLI stores the token as a generic Bearer secret keyed to the URL host,
+  // so the container only ever sees ANTHROPIC_BASE_URL + a placeholder.
+  const customBaseUrl = process.env.NANOCLAW_ANTHROPIC_BASE_URL?.trim();
+  const customAuthToken = process.env.NANOCLAW_ANTHROPIC_AUTH_TOKEN?.trim();
+  if (customBaseUrl && customAuthToken) {
+    await runCustomEndpointAuth(customBaseUrl, customAuthToken);
+    return;
+  }
+
   const method = ensureAnswer(
     await brightSelect({
       message: 'How would you like to connect to Claude?',
@@ -573,15 +680,11 @@ async function runAuthStep(): Promise<void> {
 }
 
 async function runSubscriptionAuth(): Promise<void> {
-  p.log.step("Opening the Claude sign-in flow…");
-  console.log(
-    k.dim('   (a browser will open for sign-in; this part is interactive)'),
-  );
+  p.log.step('Opening the Claude sign-in flow…');
+  console.log(k.dim('   (a browser will open for sign-in; this part is interactive)'));
   console.log();
   const start = Date.now();
-  const code = await runInheritScript('bash', [
-    'setup/register-claude-token.sh',
-  ]);
+  const code = await runInheritScript('bash', ['setup/register-claude-token.sh']);
   const durationMs = Date.now() - start;
   console.log();
   if (code !== 0) {
@@ -621,11 +724,16 @@ async function runPasteAuth(method: 'oauth' | 'api'): Promise<void> {
     'auth',
     'onecli',
     [
-      'secrets', 'create',
-      '--name', 'Anthropic',
-      '--type', 'anthropic',
-      '--value', token,
-      '--host-pattern', 'api.anthropic.com',
+      'secrets',
+      'create',
+      '--name',
+      'Anthropic',
+      '--type',
+      'anthropic',
+      '--value',
+      token,
+      '--host-pattern',
+      'api.anthropic.com',
     ],
     {
       running: `Saving your ${label} to your OneCLI vault…`,
@@ -644,6 +752,92 @@ async function runPasteAuth(method: 'oauth' | 'api'): Promise<void> {
   }
 }
 
+/**
+ * Set up Anthropic auth for a custom endpoint. The token is stored as a
+ * OneCLI generic secret with header injection so the proxy rewrites the
+ * Authorization header on the wire — the container only ever sees
+ * ANTHROPIC_BASE_URL + a placeholder bearer.
+ */
+async function runCustomEndpointAuth(
+  baseUrl: string,
+  token: string,
+): Promise<void> {
+  let host: string;
+  try {
+    host = new URL(baseUrl).hostname;
+  } catch {
+    await fail(
+      'auth',
+      `Invalid Anthropic base URL: ${baseUrl}`,
+      'Check --anthropic-base-url and retry.',
+    );
+    return;
+  }
+
+  const res = await runQuietChild(
+    'auth',
+    'onecli',
+    [
+      'secrets',
+      'create',
+      '--name',
+      'Anthropic',
+      '--type',
+      'generic',
+      '--value',
+      token,
+      '--host-pattern',
+      host,
+      '--header-name',
+      'Authorization',
+      '--value-format',
+      'Bearer {value}',
+    ],
+    {
+      running: `Saving your Anthropic auth token to your OneCLI vault…`,
+      done: 'Claude account connected.',
+    },
+    { extraFields: { METHOD: 'custom-endpoint', HOST: host } },
+  );
+  if (!res.ok) {
+    await fail(
+      'auth',
+      `Couldn't save your Anthropic auth token to the vault.`,
+      'Make sure OneCLI is running (`onecli version`), then retry.',
+    );
+  }
+
+  // ANTHROPIC_BASE_URL has to be in .env so the runtime provider config
+  // reads it when building container env. The token is *not* written —
+  // OneCLI holds it.
+  writeEnvLine('ANTHROPIC_BASE_URL', baseUrl);
+
+  // Register the claude provider so the runtime passes ANTHROPIC_BASE_URL
+  // and the placeholder bearer into the container. Only appended when the
+  // user has configured a custom endpoint; standard installs don't load
+  // the file at all.
+  appendProviderImport('./claude.js');
+}
+
+function writeEnvLine(key: string, value: string): void {
+  const envFile = path.join(process.cwd(), '.env');
+  const content = fs.existsSync(envFile) ? fs.readFileSync(envFile, 'utf-8') : '';
+  const re = new RegExp(`^${key}=.*$`, 'm');
+  const next = re.test(content)
+    ? content.replace(re, `${key}=${value}`)
+    : content.trimEnd() + (content ? '\n' : '') + `${key}=${value}\n`;
+  fs.writeFileSync(envFile, next);
+}
+
+function appendProviderImport(modulePath: string): void {
+  const file = path.join(process.cwd(), 'src', 'providers', 'index.ts');
+  const content = fs.existsSync(file) ? fs.readFileSync(file, 'utf-8') : '';
+  const line = `import '${modulePath}';`;
+  if (content.includes(line)) return;
+  const sep = content && !content.endsWith('\n') ? '\n' : '';
+  fs.writeFileSync(file, content + sep + line + '\n');
+}
+
 // ─── timezone step ─────────────────────────────────────────────────────
 
 /**
@@ -664,10 +858,7 @@ async function runTimezoneStep(): Promise<void> {
   const fields = res.terminal?.fields ?? {};
   const resolvedTz = fields.RESOLVED_TZ;
   const needsInput = fields.NEEDS_USER_INPUT === 'true';
-  const isUtc =
-    resolvedTz === 'UTC' ||
-    resolvedTz === 'Etc/UTC' ||
-    resolvedTz === 'Universal';
+  const isUtc = resolvedTz === 'UTC' || resolvedTz === 'Etc/UTC' || resolvedTz === 'Universal';
 
   // Three branches:
   //   - no TZ detected: ask where they are (or leave as UTC)
@@ -689,8 +880,8 @@ async function runTimezoneStep(): Promise<void> {
   const message = needsInput
     ? "Your system didn't expose a timezone. Which one are you in?"
     : !isUtc
-      ? "Where are you, then?"
-      : "Your system reports UTC as the timezone. Is that right, or are you somewhere else?";
+      ? 'Where are you, then?'
+      : 'Your system reports UTC as the timezone. Is that right, or are you somewhere else?';
 
   // For the non-UTC "detected-but-wrong" branch we skip the select and jump
   // straight to the free-text prompt — the user already said "not that".
@@ -717,7 +908,7 @@ async function runTimezoneStep(): Promise<void> {
 
   const answer = ensureAnswer(
     await p.text({
-      message: "Where are you? (city, region, or IANA zone)",
+      message: 'Where are you? (city, region, or IANA zone)',
       placeholder: 'e.g. New York, London, Asia/Tokyo',
       validate: (v) => (v && v.trim() ? undefined : 'Required'),
     }),
@@ -786,16 +977,30 @@ async function askDisplayName(fallback: string): Promise<string> {
   return value;
 }
 
-async function askChannelChoice(): Promise<
-  'telegram' | 'discord' | 'whatsapp' | 'teams' | 'skip'
-> {
+async function askChannelChoice(): Promise<ChannelChoice> {
+  const isMac = process.platform === 'darwin';
   const choice = ensureAnswer(
-    await brightSelect({
+    await brightSelect<ChannelChoice>({
       message: 'Want to chat with your assistant from your phone?',
       options: [
         { value: 'telegram', label: 'Yes, connect Telegram', hint: 'recommended' },
         { value: 'discord', label: 'Yes, connect Discord' },
         { value: 'whatsapp', label: 'Yes, connect WhatsApp' },
+        {
+          value: 'signal',
+          label: 'Yes, connect Signal',
+          hint: 'needs signal-cli installed',
+        },
+        {
+          value: 'imessage',
+          label: 'Yes, connect iMessage (experimental)',
+          hint: isMac ? 'local macOS mode' : 'remote Photon only',
+        },
+        {
+          value: 'slack',
+          label: 'Yes, connect Slack (experimental)',
+          hint: 'needs public URL',
+        },
         { value: 'teams', label: 'Yes, connect Microsoft Teams', hint: 'complex setup' },
         { value: 'skip', label: 'Skip for now', hint: "I'll just use the terminal" },
       ],
@@ -803,7 +1008,7 @@ async function askChannelChoice(): Promise<
   );
   setupLog.userInput('channel_choice', String(choice));
   phEmit('channel_chosen', { channel: String(choice) });
-  return choice as 'telegram' | 'discord' | 'whatsapp' | 'teams' | 'skip';
+  return choice;
 }
 
 // ─── interactive / env helpers ─────────────────────────────────────────
@@ -899,17 +1104,15 @@ function printIntro(): void {
   const wordmark = `${k.bold('Nano')}${brandBold('Claw')}`;
 
   if (isReexec) {
-    p.intro(
-      `${brandChip(' Welcome ')}  ${wordmark}  ${k.dim('· picking up where we left off')}`,
-    );
+    p.intro(`${brandChip(' Welcome ')}  ${wordmark}  ${k.dim('· picking up where we left off')}`);
     return;
   }
 
-  // Always include the wordmark inside the clack intro line. When bash ran
-  // first (NANOCLAW_BOOTSTRAPPED=1) it already printed its own wordmark
-  // above us; the small repeat is worth it to keep the brand anchored at
-  // the visible top of the clack session once the bash output scrolls away.
-  p.intro(`${wordmark}  ${k.dim("Let's get you set up.")}`);
+  // bash already printed the wordmark above us; the clack intro carries the
+  // welcome framing alone so the two don't double up. Standalone runs of
+  // setup:auto still see this as the first line — fine without the wordmark
+  // since the line itself signals the start of the flow.
+  p.intro("Let's get you set up.");
 }
 
 /**

setup/channels/imessage.ts

@@ -0,0 +1,314 @@
+/**
+ * iMessage channel flow for setup:auto.
+ *
+ * `runIMessageChannel(displayName)` covers both deployment modes:
+ *
+ *   Local (macOS): the bot runs on this Mac and talks via the signed-in
+ *   iMessage account. Reading chat.db needs Full Disk Access granted to
+ *   the Node binary — we open the directory for them so they can drag
+ *   the `node` file into System Settings.
+ *
+ *   Remote (Photon API): the bot talks to a separate server (Photon)
+ *   that owns an iMessage account on another Mac. Used when this host
+ *   is Linux, or when the operator wants to keep their daily-driver
+ *   Mac's chat history out of the loop.
+ *
+ * Flow:
+ *   1. Pick mode (auto-defaults to local on macOS, remote elsewhere)
+ *   2. Local: FDA walkthrough (open node bin directory, wait for ack)
+ *      Remote: prompt for Photon server URL + API key
+ *   3. Ask for the phone or email the operator messages from — this is
+ *      the platform-id for first-agent wiring
+ *   4. Install the adapter (setup/add-imessage.sh, non-interactive)
+ *   5. Wire the agent via scripts/init-first-agent.ts — the welcome
+ *      iMessage goes out through the normal delivery path
+ *
+ * All output obeys the three-level contract. See docs/setup-flow.md.
+ */
+import { execSync } from 'child_process';
+import os from 'os';
+import path from 'path';
+
+import * as p from '@clack/prompts';
+import k from 'kleur';
+
+import * as setupLog from '../logs.js';
+import { brightSelect } from '../lib/bright-select.js';
+import { askOperatorRole } from '../lib/role-prompt.js';
+import { ensureAnswer, fail, runQuietChild } from '../lib/runner.js';
+import { wrapForGutter } from '../lib/theme.js';
+
+const DEFAULT_AGENT_NAME = 'Nano';
+
+type Mode = 'local' | 'remote';
+
+interface RemoteCreds {
+  serverUrl: string;
+  apiKey: string;
+}
+
+export async function runIMessageChannel(displayName: string): Promise<void> {
+  const isMac = os.platform() === 'darwin';
+
+  const mode = await askMode(isMac);
+  let remoteCreds: RemoteCreds | null = null;
+
+  if (mode === 'local') {
+    if (!isMac) {
+      await fail(
+        'imessage',
+        "Local iMessage mode only works on macOS.",
+        'Choose remote mode (Photon API) on Linux/WSL, or run setup from your Mac.',
+      );
+    }
+    await walkThroughFullDiskAccess();
+  } else {
+    remoteCreds = await collectRemoteCreds();
+  }
+
+  const handle = await askOperatorHandle();
+
+  const install = await runQuietChild(
+    'imessage-install',
+    'bash',
+    ['setup/add-imessage.sh'],
+    {
+      running:
+        mode === 'local'
+          ? "Connecting the iMessage adapter to this Mac…"
+          : `Connecting the iMessage adapter to ${remoteCreds!.serverUrl}…`,
+      done: 'iMessage adapter installed.',
+    },
+    {
+      env:
+        mode === 'local'
+          ? { IMESSAGE_LOCAL: 'true', IMESSAGE_ENABLED: 'true' }
+          : {
+              IMESSAGE_LOCAL: 'false',
+              IMESSAGE_SERVER_URL: remoteCreds!.serverUrl,
+              IMESSAGE_API_KEY: remoteCreds!.apiKey,
+            },
+      extraFields: { MODE: mode },
+    },
+  );
+  if (!install.ok) {
+    await fail(
+      'imessage-install',
+      "Couldn't install the iMessage adapter.",
+      'See logs/setup-steps/ for details, then retry setup.',
+    );
+  }
+
+  const role = await askOperatorRole('iMessage');
+  setupLog.userInput('imessage_role', role);
+
+  const agentName = await resolveAgentName();
+
+  const init = await runQuietChild(
+    'init-first-agent',
+    'pnpm',
+    [
+      'exec', 'tsx', 'scripts/init-first-agent.ts',
+      '--channel', 'imessage',
+      '--user-id', handle,
+      '--platform-id', handle,
+      '--display-name', displayName,
+      '--agent-name', agentName,
+      '--role', role,
+    ],
+    {
+      running: `Connecting ${agentName} to iMessage…`,
+      done: `${agentName} is ready. Check iMessage for a welcome message.`,
+    },
+    {
+      extraFields: {
+        CHANNEL: 'imessage',
+        AGENT_NAME: agentName,
+        PLATFORM_ID: handle,
+        MODE: mode,
+      },
+    },
+  );
+  if (!init.ok) {
+    await fail(
+      'init-first-agent',
+      `Couldn't finish connecting ${agentName}.`,
+      'Double-check Full Disk Access (local mode) or Photon credentials (remote), then retry.',
+    );
+  }
+}
+
+async function askMode(isMac: boolean): Promise<Mode> {
+  const choice = ensureAnswer(
+    await brightSelect<Mode>({
+      message: 'How should iMessage run?',
+      initialValue: isMac ? 'local' : 'remote',
+      options: isMac
+        ? [
+            {
+              value: 'local',
+              label: 'Local (this Mac)',
+              hint: "uses this machine's iMessage account",
+            },
+            {
+              value: 'remote',
+              label: 'Remote (Photon API)',
+              hint: 'the bot lives on another server',
+            },
+          ]
+        : [
+            {
+              value: 'remote',
+              label: 'Remote (Photon API)',
+              hint: 'only option off macOS',
+            },
+          ],
+    }),
+  );
+  setupLog.userInput('imessage_mode', String(choice));
+  return choice;
+}
+
+/**
+ * Grant Full Disk Access to the Node binary the host runs under — without
+ * it, the adapter can't read chat.db and inbound messages never arrive.
+ * Opening the containing directory in Finder makes the drag-and-drop
+ * target obvious; falling back to printing the path keeps us working in
+ * SSH/headless contexts where `open` is a no-op.
+ */
+async function walkThroughFullDiskAccess(): Promise<void> {
+  let nodePath = process.execPath;
+  try {
+    // `which node` picks up the user's shell-resolved node, which may differ
+    // from process.execPath (e.g. they launched setup under a different
+    // Node via `nvm`). If it succeeds and is resolvable, prefer it.
+    const which = execSync('which node', { encoding: 'utf-8' }).trim();
+    if (which) nodePath = which;
+  } catch {
+    // fall back to process.execPath
+  }
+  const nodeDir = path.dirname(nodePath);
+
+  p.note(
+    wrapForGutter(
+      [
+        `iMessage needs Full Disk Access granted to the Node binary:`,
+        '',
+        `  ${nodePath}`,
+        '',
+        '  1. System Settings → Privacy & Security → Full Disk Access',
+        `  2. Click +, then drag the "node" file from the Finder window`,
+        '     we just opened for you',
+        '  3. Toggle it on, then come back here',
+      ].join('\n'),
+      6,
+    ),
+    'Grant Full Disk Access',
+  );
+
+  try {
+    execSync(`open "${nodeDir}"`, { stdio: 'ignore' });
+  } catch {
+    // No Finder (SSH/headless) — user sees the path in the note above.
+  }
+
+  ensureAnswer(
+    await p.confirm({
+      message: "Granted Full Disk Access?",
+      initialValue: true,
+    }),
+  );
+  setupLog.userInput('imessage_fda_confirmed', 'true');
+}
+
+async function collectRemoteCreds(): Promise<RemoteCreds> {
+  p.note(
+    [
+      "Photon is a separate service that owns an iMessage account and",
+      "exposes it over HTTP. NanoClaw will talk to it via its API.",
+      '',
+      '  1. Set up a Photon server: https://photon.im',
+      '  2. Copy the server URL and API key from your Photon dashboard',
+    ].join('\n'),
+    'Remote iMessage via Photon',
+  );
+
+  const urlAnswer = ensureAnswer(
+    await p.text({
+      message: 'Photon server URL',
+      placeholder: 'https://photon.example.com',
+      validate: (v) => {
+        const t = (v ?? '').trim();
+        if (!t) return 'URL is required';
+        if (!/^https?:\/\//i.test(t)) return 'Must start with http:// or https://';
+        return undefined;
+      },
+    }),
+  );
+  const serverUrl = (urlAnswer as string).trim();
+
+  const keyAnswer = ensureAnswer(
+    await p.password({
+      message: 'Photon API key',
+      validate: (v) => ((v ?? '').trim() ? undefined : 'API key is required'),
+    }),
+  );
+  const apiKey = (keyAnswer as string).trim();
+
+  setupLog.userInput('imessage_server_url', serverUrl);
+  setupLog.userInput(
+    'imessage_api_key',
+    `${apiKey.slice(0, 4)}…${apiKey.slice(-4)}`,
+  );
+  return { serverUrl, apiKey };
+}
+
+async function askOperatorHandle(): Promise<string> {
+  p.note(
+    [
+      "What phone number or email do you iMessage with?",
+      "That's where your assistant will send its welcome message.",
+      '',
+      k.dim('  • Phone: full E.164, e.g. +15551234567'),
+      k.dim('  • Email: whatever iMessage recognises (Apple ID, iCloud alias, …)'),
+    ].join('\n'),
+    'Your iMessage handle',
+  );
+
+  const answer = ensureAnswer(
+    await p.text({
+      message: 'Phone number or email',
+      validate: (v) => {
+        const t = (v ?? '').trim();
+        if (!t) return 'Required';
+        const isPhone = /^\+\d{8,15}$/.test(t);
+        const isEmail = /^[^\s@]+@[^\s@]+\.[^\s@]+$/.test(t);
+        if (!isPhone && !isEmail) {
+          return "Use a +E.164 phone number or an email address";
+        }
+        return undefined;
+      },
+    }),
+  );
+  const handle = (answer as string).trim();
+  setupLog.userInput('imessage_handle', handle);
+  return handle;
+}
+
+async function resolveAgentName(): Promise<string> {
+  const preset = process.env.NANOCLAW_AGENT_NAME?.trim();
+  if (preset) {
+    setupLog.userInput('agent_name', preset);
+    return preset;
+  }
+  const answer = ensureAnswer(
+    await p.text({
+      message: 'What should your assistant be called?',
+      placeholder: DEFAULT_AGENT_NAME,
+      defaultValue: DEFAULT_AGENT_NAME,
+    }),
+  );
+  const value = (answer as string).trim() || DEFAULT_AGENT_NAME;
+  setupLog.userInput('agent_name', value);
+  return value;
+}

setup/channels/signal.ts

@@ -0,0 +1,357 @@
+/**
+ * Signal channel flow for setup:auto.
+ *
+ * `runSignalChannel(displayName)` owns the full branch from signal-cli
+ * presence check through the welcome DM:
+ *
+ *   1. Probe signal-cli on PATH (or SIGNAL_CLI_PATH). On macOS without it,
+ *      offer `brew install signal-cli` inline. On Linux, surface the
+ *      GitHub releases URL and bail with an actionable error.
+ *   2. Install the adapter + qrcode via setup/add-signal.sh (idempotent).
+ *   3. Run the signal-auth step, rendering each SIGNAL_AUTH_QR block as
+ *      a terminal QR the operator scans from Signal → Linked Devices.
+ *   4. Persist SIGNAL_ACCOUNT to .env (+ data/env/env).
+ *   5. Kick the service so the adapter picks up the new credentials.
+ *   6. Ask operator role + agent name.
+ *   7. Wire the agent via scripts/init-first-agent.ts; the existing welcome
+ *      DM path delivers the greeting through the adapter.
+ *
+ * Signal's `link` flow creates a *secondary* device. The phone number
+ * comes from the primary (the phone that scanned the QR); this host then
+ * sends/receives as that primary number. No registration of new numbers.
+ *
+ * Output obeys the three-level contract: clack UI for the user, structured
+ * entries in logs/setup.log, full raw output in per-step files under
+ * logs/setup-steps/. See docs/setup-flow.md.
+ */
+import { spawnSync } from 'child_process';
+import fs from 'fs';
+import path from 'path';
+
+import * as p from '@clack/prompts';
+import k from 'kleur';
+
+import * as setupLog from '../logs.js';
+import { getLaunchdLabel, getSystemdUnit } from '../../src/install-slug.js';
+import {
+  type Block,
+  type StepResult,
+  dumpTranscriptOnFailure,
+  ensureAnswer,
+  fail,
+  runQuietChild,
+  spawnStep,
+  writeStepEntry,
+} from '../lib/runner.js';
+import { askOperatorRole } from '../lib/role-prompt.js';
+
+const DEFAULT_AGENT_NAME = 'Nano';
+
+export async function runSignalChannel(displayName: string): Promise<void> {
+  await ensureSignalCli();
+
+  const install = await runQuietChild(
+    'signal-install',
+    'bash',
+    ['setup/add-signal.sh'],
+    {
+      running: 'Installing the Signal adapter…',
+      done: 'Signal adapter installed.',
+      skipped: 'Signal adapter already installed.',
+    },
+  );
+  if (!install.ok) {
+    await fail(
+      'signal-install',
+      "Couldn't install the Signal adapter.",
+      'See logs/setup-steps/ for details, then retry setup.',
+    );
+  }
+
+  const auth = await runSignalAuth();
+  if (!auth.ok) {
+    const reason = auth.terminal?.fields.ERROR ?? 'unknown';
+    await fail(
+      'signal-auth',
+      `Signal link failed (${reason}).`,
+      reason === 'qr_timeout'
+        ? 'The code expired. Re-run setup to get a fresh one.'
+        : 'Re-run setup to try again.',
+    );
+  }
+
+  const account = auth.terminal?.fields.ACCOUNT;
+  if (!account) {
+    await fail(
+      'signal-auth',
+      'Linked with Signal but couldn\'t read the phone number back.',
+      'Run `signal-cli listAccounts` to confirm, then re-run setup.',
+    );
+  }
+
+  writeSignalAccount(account!);
+  await restartService();
+
+  const role = await askOperatorRole('Signal');
+  setupLog.userInput('signal_role', role);
+
+  const agentName = await resolveAgentName();
+
+  const init = await runQuietChild(
+    'init-first-agent',
+    'pnpm',
+    [
+      'exec', 'tsx', 'scripts/init-first-agent.ts',
+      '--channel', 'signal',
+      '--user-id', account!,
+      '--platform-id', account!,
+      '--display-name', displayName,
+      '--agent-name', agentName,
+      '--role', role,
+    ],
+    {
+      running: `Connecting ${agentName} to Signal…`,
+      done: `${agentName} is ready. Check Signal for a welcome message.`,
+    },
+    {
+      extraFields: {
+        CHANNEL: 'signal',
+        AGENT_NAME: agentName,
+        PLATFORM_ID: account!,
+        ROLE: role,
+      },
+    },
+  );
+  if (!init.ok) {
+    await fail(
+      'init-first-agent',
+      `Couldn't finish connecting ${agentName}.`,
+      'You can retry later with `/manage-channels`.',
+    );
+  }
+}
+
+async function ensureSignalCli(): Promise<void> {
+  const cli = process.env.SIGNAL_CLI_PATH || 'signal-cli';
+  const probe = spawnSync(cli, ['--version'], {
+    stdio: ['ignore', 'pipe', 'pipe'],
+  });
+  if (!probe.error && probe.status === 0) return;
+
+  if (process.platform === 'darwin') {
+    p.note(
+      [
+        "NanoClaw talks to Signal through signal-cli, which isn't installed yet.",
+        '',
+        'The quickest way on macOS is Homebrew:',
+        '',
+        k.cyan('  brew install signal-cli'),
+        '',
+        "Install it in another terminal, then re-run setup.",
+      ].join('\n'),
+      'signal-cli not found',
+    );
+  } else {
+    p.note(
+      [
+        "NanoClaw talks to Signal through signal-cli, which isn't installed yet.",
+        '',
+        'Grab the latest release from GitHub:',
+        '',
+        k.cyan('  https://github.com/AsamK/signal-cli/releases'),
+        '',
+        "Install it, make sure `signal-cli --version` works, then re-run setup.",
+      ].join('\n'),
+      'signal-cli not found',
+    );
+  }
+  await fail(
+    'signal-install',
+    'signal-cli is required but not installed.',
+    'Install it and re-run setup.',
+  );
+}
+
+async function runSignalAuth(): Promise<
+  StepResult & { rawLog: string; durationMs: number }
+> {
+  const rawLog = setupLog.stepRawLog('signal-auth');
+  const start = Date.now();
+  const s = p.spinner();
+  s.start('Starting Signal link…');
+  let spinnerActive = true;
+
+  const stopSpinner = (msg: string, code?: number): void => {
+    if (spinnerActive) {
+      s.stop(msg, code);
+      spinnerActive = false;
+    }
+  };
+
+  // Tracks how many lines the QR block occupies so we can wipe it in-place
+  // once linking succeeds (Signal's link URL doesn't rotate like WhatsApp's,
+  // but we still want to erase the QR from screen once it's served).
+  let qrLinesPrinted = 0;
+
+  const result = await spawnStep(
+    'signal-auth',
+    [],
+    (block: Block) => {
+      if (block.type === 'SIGNAL_AUTH_QR') {
+        const qr = block.fields.QR ?? '';
+        if (!qr) return;
+        void renderQr(qr).then((lines) => {
+          stopSpinner('Scan this QR from Signal → Settings → Linked Devices.');
+          process.stdout.write(lines.join('\n') + '\n');
+          qrLinesPrinted = lines.length;
+          s.start('Waiting for you to scan…');
+          spinnerActive = true;
+        });
+      } else if (block.type === 'SIGNAL_AUTH') {
+        const status = block.fields.STATUS;
+        // Wipe the QR block regardless of outcome — it's either scanned
+        // and useless, or expired and misleading.
+        if (qrLinesPrinted > 0) {
+          process.stdout.write(`\x1b[${qrLinesPrinted}A\x1b[0J`);
+          qrLinesPrinted = 0;
+        }
+        const account = block.fields.ACCOUNT;
+        if (status === 'skipped') {
+          stopSpinner(
+            account
+              ? `Signal already linked as ${k.cyan(account)}.`
+              : 'Signal already linked.',
+          );
+        } else if (status === 'success') {
+          stopSpinner(`Signal linked as ${k.cyan(String(account ?? ''))}.`);
+        } else if (status === 'failed') {
+          const err = block.fields.ERROR ?? 'unknown';
+          stopSpinner(`Signal link failed: ${err}`, 1);
+        }
+      }
+    },
+    rawLog,
+  );
+  const durationMs = Date.now() - start;
+
+  if (spinnerActive) {
+    stopSpinner(
+      result.ok ? 'Done.' : 'Signal link ended unexpectedly.',
+      result.ok ? 0 : 1,
+    );
+    if (!result.ok) dumpTranscriptOnFailure(result.transcript);
+  }
+
+  writeStepEntry('signal-auth', result, durationMs, rawLog);
+  return { ...result, rawLog, durationMs };
+}
+
+/**
+ * Render the raw linking URL as a block-art QR, returned line-by-line so
+ * the caller can count lines for in-place cleanup. Uses small-mode so the
+ * code stays scannable on 24-row terminals. If qrcode isn't installed
+ * (add-signal.sh should have handled it, but we're defensive), fall back
+ * to the raw URL and ask the user to paste it into an external renderer.
+ */
+async function renderQr(url: string): Promise<string[]> {
+  try {
+    const QRCode = await import('qrcode');
+    const qrText = await QRCode.toString(url, { type: 'terminal', small: true });
+    const caption = k.dim(
+      '   Signal → Settings → Linked Devices → Link New Device → scan.',
+    );
+    return [...qrText.trimEnd().split('\n'), '', caption];
+  } catch {
+    return [
+      'Linking URL (render at https://qr.io or similar):',
+      '',
+      url,
+      '',
+      k.dim('Signal → Settings → Linked Devices → Link New Device → scan.'),
+    ];
+  }
+}
+
+/** Persist SIGNAL_ACCOUNT to .env and mirror to data/env/env for the container. */
+function writeSignalAccount(account: string): void {
+  const envPath = path.join(process.cwd(), '.env');
+  let contents = '';
+  try {
+    contents = fs.readFileSync(envPath, 'utf-8');
+  } catch {
+    contents = '';
+  }
+  if (/^SIGNAL_ACCOUNT=/m.test(contents)) {
+    contents = contents.replace(
+      /^SIGNAL_ACCOUNT=.*$/m,
+      `SIGNAL_ACCOUNT=${account}`,
+    );
+  } else {
+    if (contents.length > 0 && !contents.endsWith('\n')) contents += '\n';
+    contents += `SIGNAL_ACCOUNT=${account}\n`;
+  }
+  fs.writeFileSync(envPath, contents);
+
+  const containerEnvDir = path.join(process.cwd(), 'data', 'env');
+  fs.mkdirSync(containerEnvDir, { recursive: true });
+  fs.copyFileSync(envPath, path.join(containerEnvDir, 'env'));
+
+  setupLog.userInput('signal_account', account);
+}
+
+async function restartService(): Promise<void> {
+  const s = p.spinner();
+  s.start('Restarting NanoClaw so it sees your Signal account…');
+  const start = Date.now();
+  const platform = process.platform;
+  try {
+    if (platform === 'darwin') {
+      spawnSync(
+        'launchctl',
+        ['kickstart', '-k', `gui/${process.getuid?.() ?? 501}/${getLaunchdLabel()}`],
+        { stdio: 'ignore' },
+      );
+    } else if (platform === 'linux') {
+      const unit = getSystemdUnit();
+      const user = spawnSync('systemctl', ['--user', 'restart', unit], {
+        stdio: 'ignore',
+      });
+      if (user.status !== 0) {
+        spawnSync('sudo', ['systemctl', 'restart', unit], { stdio: 'ignore' });
+      }
+    }
+    // Give the adapter a moment to connect to signal-cli before
+    // init-first-agent's welcome DM hits the delivery path.
+    await new Promise((r) => setTimeout(r, 5000));
+    const elapsed = Math.round((Date.now() - start) / 1000);
+    s.stop(`NanoClaw restarted. ${k.dim(`(${elapsed}s)`)}`);
+    setupLog.step('signal-restart', 'success', Date.now() - start, {
+      PLATFORM: platform,
+    });
+  } catch (err) {
+    const message = err instanceof Error ? err.message : String(err);
+    s.stop(`Restart may have failed: ${message}`, 1);
+    setupLog.step('signal-restart', 'failed', Date.now() - start, {
+      ERROR: message,
+    });
+    // Non-fatal — the user can restart manually if init-first-agent fails.
+  }
+}
+
+async function resolveAgentName(): Promise<string> {
+  const preset = process.env.NANOCLAW_AGENT_NAME?.trim();
+  if (preset) {
+    setupLog.userInput('agent_name', preset);
+    return preset;
+  }
+  const answer = ensureAnswer(
+    await p.text({
+      message: 'What should your assistant be called?',
+      placeholder: DEFAULT_AGENT_NAME,
+      defaultValue: DEFAULT_AGENT_NAME,
+    }),
+  );
+  const value = (answer as string).trim() || DEFAULT_AGENT_NAME;
+  setupLog.userInput('agent_name', value);
+  return value;
+}

setup/channels/slack.ts

@@ -0,0 +1,249 @@
+/**
+ * Slack channel flow for setup:auto.
+ *
+ * `runSlackChannel(displayName)` walks the operator from a bare Slack
+ * workspace through a running bot, then stops before wiring an agent:
+ *
+ *   1. Walk through creating a Slack app (api.slack.com/apps) — scopes,
+ *      event subscriptions, and signing secret
+ *   2. Paste the bot token + signing secret (clack password prompts)
+ *   3. Validate via auth.test → resolves workspace + bot identity
+ *   4. Install the adapter (setup/add-slack.sh, non-interactive)
+ *   5. Print the post-install checklist: set the public webhook URL in
+ *      Slack's Event Subscriptions, DM the bot to bootstrap the channel,
+ *      then `/manage-channels` to wire an agent.
+ *
+ * Why no welcome DM here: unlike Discord/Telegram (gateway / long-poll),
+ * Slack needs a public Event Subscriptions URL for inbound events, and
+ * opening an unsolicited DM would need `im:write` scope we don't force
+ * the SKILL.md to require. Shipping a honest "here's what's left" note
+ * is better than a welcome DM the user won't receive until they
+ * configure the webhook anyway.
+ *
+ * All output obeys the three-level contract. See docs/setup-flow.md.
+ */
+import * as p from '@clack/prompts';
+import k from 'kleur';
+
+import * as setupLog from '../logs.js';
+import { confirmThenOpen } from '../lib/browser.js';
+import { ensureAnswer, fail, runQuietChild } from '../lib/runner.js';
+import { wrapForGutter } from '../lib/theme.js';
+
+const SLACK_API = 'https://slack.com/api';
+const SLACK_APPS_URL = 'https://api.slack.com/apps';
+
+interface WorkspaceInfo {
+  teamName: string;
+  teamId: string;
+  botName: string;
+  botUserId: string;
+}
+
+// displayName is reserved for when we start wiring the first agent here.
+// Kept to match the `run<X>Channel(displayName)` signature every other
+// channel driver uses, so auto.ts can dispatch without a branch.
+export async function runSlackChannel(_displayName: string): Promise<void> {
+  await walkThroughAppCreation();
+
+  const token = await collectBotToken();
+  const signingSecret = await collectSigningSecret();
+  const info = await validateSlackToken(token);
+
+  const install = await runQuietChild(
+    'slack-install',
+    'bash',
+    ['setup/add-slack.sh'],
+    {
+      running: `Connecting Slack to @${info.botName} (${info.teamName})…`,
+      done: 'Slack adapter installed.',
+    },
+    {
+      env: {
+        SLACK_BOT_TOKEN: token,
+        SLACK_SIGNING_SECRET: signingSecret,
+      },
+      extraFields: {
+        BOT_NAME: info.botName,
+        TEAM_NAME: info.teamName,
+        TEAM_ID: info.teamId,
+      },
+    },
+  );
+  if (!install.ok) {
+    await fail(
+      'slack-install',
+      "Couldn't connect Slack.",
+      'See logs/setup-steps/ for details, then retry setup.',
+    );
+  }
+
+  showPostInstallChecklist(info);
+}
+
+async function walkThroughAppCreation(): Promise<void> {
+  p.note(
+    [
+      "You'll create a Slack app that the assistant talks through.",
+      "Free and stays inside the workspaces you pick.",
+      '',
+      '  1. Create a new app "From scratch", name it, pick a workspace',
+      '  2. OAuth & Permissions → add Bot Token Scopes:',
+      '     chat:write, channels:history, groups:history, im:history,',
+      '     channels:read, groups:read, users:read, reactions:write',
+      '  3. App Home → enable "Messages Tab" and "Allow users to send',
+      '     slash commands and messages from the messages tab"',
+      '  4. Basic Information → copy the "Signing Secret"',
+      '  5. Install to Workspace → copy the "Bot User OAuth Token" (xoxb-…)',
+      '',
+      k.dim(SLACK_APPS_URL),
+    ].join('\n'),
+    'Create a Slack app',
+  );
+  await confirmThenOpen(SLACK_APPS_URL, 'Press Enter to open Slack app settings');
+
+  ensureAnswer(
+    await p.confirm({
+      message: 'Got your bot token and signing secret?',
+      initialValue: true,
+    }),
+  );
+}
+
+async function collectBotToken(): Promise<string> {
+  const answer = ensureAnswer(
+    await p.password({
+      message: 'Paste your Slack bot token',
+      validate: (v) => {
+        const t = (v ?? '').trim();
+        if (!t) return 'Token is required';
+        if (!t.startsWith('xoxb-')) return 'Bot tokens start with xoxb-';
+        if (t.length < 24) return "That's shorter than a real Slack bot token";
+        return undefined;
+      },
+    }),
+  );
+  const token = (answer as string).trim();
+  setupLog.userInput(
+    'slack_bot_token',
+    `${token.slice(0, 10)}…${token.slice(-4)}`,
+  );
+  return token;
+}
+
+async function collectSigningSecret(): Promise<string> {
+  const answer = ensureAnswer(
+    await p.password({
+      message: 'Paste your Slack signing secret',
+      validate: (v) => {
+        const t = (v ?? '').trim();
+        if (!t) return 'Signing secret is required';
+        // Slack signing secrets are 32-char hex strings, but newer apps
+        // sometimes emit longer variants — leniently require hex only.
+        if (!/^[a-f0-9]{16,}$/i.test(t)) {
+          return 'Signing secrets are a string of hex characters';
+        }
+        return undefined;
+      },
+    }),
+  );
+  const secret = (answer as string).trim();
+  setupLog.userInput(
+    'slack_signing_secret',
+    `${secret.slice(0, 4)}…${secret.slice(-4)}`,
+  );
+  return secret;
+}
+
+async function validateSlackToken(token: string): Promise<WorkspaceInfo> {
+  const s = p.spinner();
+  const start = Date.now();
+  s.start('Checking your bot token…');
+  try {
+    const res = await fetch(`${SLACK_API}/auth.test`, {
+      method: 'POST',
+      headers: {
+        Authorization: `Bearer ${token}`,
+        'Content-Type': 'application/x-www-form-urlencoded',
+      },
+    });
+    const data = (await res.json()) as {
+      ok?: boolean;
+      team?: string;
+      team_id?: string;
+      user?: string;
+      user_id?: string;
+      error?: string;
+    };
+    const elapsedS = Math.round((Date.now() - start) / 1000);
+    if (data.ok && data.team && data.user) {
+      s.stop(
+        `Connected to ${data.team} as @${data.user}. ${k.dim(`(${elapsedS}s)`)}`,
+      );
+      const info: WorkspaceInfo = {
+        teamName: data.team,
+        teamId: data.team_id ?? '',
+        botName: data.user,
+        botUserId: data.user_id ?? '',
+      };
+      setupLog.step('slack-validate', 'success', Date.now() - start, {
+        BOT_NAME: info.botName,
+        BOT_USER_ID: info.botUserId,
+        TEAM_NAME: info.teamName,
+        TEAM_ID: info.teamId,
+      });
+      return info;
+    }
+    const reason = data.error ?? `HTTP ${res.status}`;
+    s.stop(`Slack didn't accept that token: ${reason}`, 1);
+    setupLog.step('slack-validate', 'failed', Date.now() - start, {
+      ERROR: reason,
+    });
+    await fail(
+      'slack-validate',
+      "Slack didn't accept that token.",
+      reason === 'invalid_auth' || reason === 'token_revoked'
+        ? 'Copy the token again from OAuth & Permissions and retry setup.'
+        : `Slack said "${reason}". Check the token scopes and workspace install, then retry.`,
+    );
+  } catch (err) {
+    const elapsedS = Math.round((Date.now() - start) / 1000);
+    s.stop(`Couldn't reach Slack. ${k.dim(`(${elapsedS}s)`)}`, 1);
+    const message = err instanceof Error ? err.message : String(err);
+    setupLog.step('slack-validate', 'failed', Date.now() - start, {
+      ERROR: message,
+    });
+    await fail(
+      'slack-validate',
+      "Couldn't reach Slack.",
+      'Check your internet connection and retry setup.',
+    );
+  }
+}
+
+function showPostInstallChecklist(info: WorkspaceInfo): void {
+  p.note(
+    wrapForGutter(
+      [
+        `The Slack adapter is installed and your creds are saved. ${info.teamName} still needs two things before it can talk to you:`,
+        '',
+        '  1. A public URL so Slack can deliver events.',
+        '     NanoClaw serves a webhook on port 3000 by default — expose it',
+        '     via ngrok, Cloudflare Tunnel, or a reverse proxy on a VPS.',
+        '',
+        '  2. In your Slack app → Event Subscriptions:',
+        '     • Toggle "Enable Events" on',
+        `     • Request URL: https://<your-public-host>/webhook/slack`,
+        '     • Subscribe to bot events: message.channels, message.groups,',
+        '       message.im, app_mention',
+        '     • Save, then reinstall the app when Slack prompts',
+        '',
+        `  3. DM @${info.botName} from Slack once — that bootstraps the`,
+        '     messaging group. Then run `/manage-channels` in `claude` to',
+        '     wire an agent to it.',
+      ].join('\n'),
+      6,
+    ),
+    'Finish setting up Slack',
+  );
+}

setup/environment.test.ts

@@ -1,5 +1,7 @@
-import { describe, it, expect, beforeEach } from 'vitest';
+import { describe, it, expect, beforeEach, afterEach } from 'vitest';
 import fs from 'fs';
+import os from 'os';
+import path from 'path';
 
 import Database from 'better-sqlite3';
 
@@ -17,58 +19,63 @@ describe('environment detection', () => {
   });
 });
 
-describe('registered groups DB query', () => {
-  let db: Database.Database;
+describe('detectRegisteredGroups', () => {
+  let tempDir: string;
 
   beforeEach(() => {
-    db = new Database(':memory:');
-    db.exec(`CREATE TABLE IF NOT EXISTS registered_groups (
-      jid TEXT PRIMARY KEY,
-      name TEXT NOT NULL,
-      folder TEXT NOT NULL UNIQUE,
-      trigger_pattern TEXT NOT NULL,
-      added_at TEXT NOT NULL,
-      container_config TEXT,
-      requires_trigger INTEGER DEFAULT 1
-    )`);
+    tempDir = fs.mkdtempSync(path.join(os.tmpdir(), 'nanoclaw-env-test-'));
+    fs.mkdirSync(path.join(tempDir, 'data'), { recursive: true });
   });
 
-  it('returns 0 for empty table', () => {
-    const row = db
-      .prepare('SELECT COUNT(*) as count FROM registered_groups')
-      .get() as { count: number };
-    expect(row.count).toBe(0);
+  afterEach(() => {
+    fs.rmSync(tempDir, { recursive: true, force: true });
   });
 
-  it('returns correct count after inserts', () => {
-    db.prepare(
-      `INSERT INTO registered_groups (jid, name, folder, trigger_pattern, added_at, requires_trigger)
-       VALUES (?, ?, ?, ?, ?, ?)`,
-    ).run(
-      '123@g.us',
-      'Group 1',
-      'group-1',
-      '@Andy',
-      '2024-01-01T00:00:00.000Z',
-      1,
-    );
+  it('returns false when no registration state exists', async () => {
+    const { detectRegisteredGroups } = await import('./environment.js');
+    expect(detectRegisteredGroups(tempDir)).toBe(false);
+  });
 
-    db.prepare(
-      `INSERT INTO registered_groups (jid, name, folder, trigger_pattern, added_at, requires_trigger)
-       VALUES (?, ?, ?, ?, ?, ?)`,
-    ).run(
-      '456@g.us',
-      'Group 2',
-      'group-2',
-      '@Andy',
-      '2024-01-01T00:00:00.000Z',
-      1,
-    );
+  it('detects pre-migration registered_groups.json', async () => {
+    const { detectRegisteredGroups } = await import('./environment.js');
+    fs.writeFileSync(path.join(tempDir, 'data', 'registered_groups.json'), '[]');
+    expect(detectRegisteredGroups(tempDir)).toBe(true);
+  });
 
-    const row = db
-      .prepare('SELECT COUNT(*) as count FROM registered_groups')
-      .get() as { count: number };
-    expect(row.count).toBe(2);
+  it('returns false for an empty v2 central DB', async () => {
+    const { detectRegisteredGroups } = await import('./environment.js');
+    const db = new Database(path.join(tempDir, 'data', 'v2.db'));
+    db.exec(`
+      CREATE TABLE agent_groups (id TEXT PRIMARY KEY);
+      CREATE TABLE messaging_group_agents (
+        id TEXT PRIMARY KEY,
+        messaging_group_id TEXT NOT NULL,
+        agent_group_id TEXT NOT NULL
+      );
+    `);
+    db.close();
+
+    expect(detectRegisteredGroups(tempDir)).toBe(false);
+  });
+
+  it('detects wired agent groups in the v2 central DB', async () => {
+    const { detectRegisteredGroups } = await import('./environment.js');
+    const db = new Database(path.join(tempDir, 'data', 'v2.db'));
+    db.exec(`
+      CREATE TABLE agent_groups (id TEXT PRIMARY KEY);
+      CREATE TABLE messaging_group_agents (
+        id TEXT PRIMARY KEY,
+        messaging_group_id TEXT NOT NULL,
+        agent_group_id TEXT NOT NULL
+      );
+    `);
+    db.prepare('INSERT INTO agent_groups (id) VALUES (?)').run('ag-1');
+    db.prepare(
+      'INSERT INTO messaging_group_agents (id, messaging_group_id, agent_group_id) VALUES (?, ?, ?)',
+    ).run('mga-1', 'mg-1', 'ag-1');
+    db.close();
+
+    expect(detectRegisteredGroups(tempDir)).toBe(true);
   });
 });
 

setup/environment.ts

@@ -7,11 +7,35 @@ import path from 'path';
 
 import Database from 'better-sqlite3';
 
-import { STORE_DIR } from '../src/config.js';
 import { log } from '../src/log.js';
 import { commandExists, getPlatform, isHeadless, isWSL } from './platform.js';
 import { emitStatus } from './status.js';
 
+export function detectRegisteredGroups(projectRoot: string): boolean {
+  if (fs.existsSync(path.join(projectRoot, 'data', 'registered_groups.json'))) {
+    return true;
+  }
+
+  const dbPath = path.join(projectRoot, 'data', 'v2.db');
+  if (!fs.existsSync(dbPath)) return false;
+
+  let db: Database.Database | null = null;
+  try {
+    db = new Database(dbPath, { readonly: true });
+    const row = db
+      .prepare(
+        `SELECT COUNT(DISTINCT ag.id) as count FROM agent_groups ag
+         JOIN messaging_group_agents mga ON mga.agent_group_id = ag.id`,
+      )
+      .get() as { count: number };
+    return row.count > 0;
+  } catch {
+    return false;
+  } finally {
+    db?.close();
+  }
+}
+
 export async function run(_args: string[]): Promise<void> {
   const projectRoot = process.cwd();
 
@@ -39,26 +63,7 @@ export async function run(_args: string[]): Promise<void> {
   const authDir = path.join(projectRoot, 'store', 'auth');
   const hasAuth = fs.existsSync(authDir) && fs.readdirSync(authDir).length > 0;
 
-  let hasRegisteredGroups = false;
-  // Check JSON file first (pre-migration)
-  if (fs.existsSync(path.join(projectRoot, 'data', 'registered_groups.json'))) {
-    hasRegisteredGroups = true;
-  } else {
-    // Check SQLite directly using better-sqlite3 (no sqlite3 CLI needed)
-    const dbPath = path.join(STORE_DIR, 'messages.db');
-    if (fs.existsSync(dbPath)) {
-      try {
-        const db = new Database(dbPath, { readonly: true });
-        const row = db
-          .prepare('SELECT COUNT(*) as count FROM registered_groups')
-          .get() as { count: number };
-        if (row.count > 0) hasRegisteredGroups = true;
-        db.close();
-      } catch {
-        // Table might not exist yet
-      }
-    }
-  }
+  const hasRegisteredGroups = detectRegisteredGroups(projectRoot);
 
   // Check for existing OpenClaw installation
   const homedir = (await import('os')).homedir();

setup/index.ts

@@ -16,6 +16,7 @@ const STEPS: Record<
   register: () => import('./register.js'),
   groups: () => import('./groups.js'),
   'whatsapp-auth': () => import('./whatsapp-auth.js'),
+  'signal-auth': () => import('./signal-auth.js'),
   mounts: () => import('./mounts.js'),
   service: () => import('./service.js'),
   verify: () => import('./verify.js'),

setup/lib/agent-ping.test.ts

@@ -0,0 +1,30 @@
+import { describe, expect, it } from 'vitest';
+
+import { classifyPingResult } from './agent-ping.js';
+
+describe('classifyPingResult', () => {
+  it('treats a normal text reply as ok', () => {
+    expect(classifyPingResult(0, 'pong\n')).toBe('ok');
+  });
+
+  it('detects Anthropic auth errors printed as a chat reply', () => {
+    expect(
+      classifyPingResult(
+        0,
+        'Failed to authenticate. API Error: 401 {"type":"error","error":{"type":"authentication_error","message":"Invalid bearer token"}}',
+      ),
+    ).toBe('auth_error');
+  });
+
+  it('detects auth errors on stderr too', () => {
+    expect(classifyPingResult(1, '', 'Authentication error')).toBe('auth_error');
+  });
+
+  it('preserves socket errors', () => {
+    expect(classifyPingResult(2, '')).toBe('socket_error');
+  });
+
+  it('treats empty output as no reply', () => {
+    expect(classifyPingResult(0, '')).toBe('no_reply');
+  });
+});

setup/lib/agent-ping.ts

@@ -13,7 +13,21 @@
  */
 import { spawn } from 'child_process';
 
-export type PingResult = 'ok' | 'no_reply' | 'socket_error';
+export type PingResult = 'ok' | 'no_reply' | 'socket_error' | 'auth_error';
+
+export function classifyPingResult(exitCode: number | null, stdout: string, stderr = ''): PingResult {
+  const output = `${stdout}\n${stderr}`;
+  if (
+    /Invalid bearer token/i.test(output) ||
+    /authentication[_ ]error/i.test(output) ||
+    /Failed to authenticate/i.test(output)
+  ) {
+    return 'auth_error';
+  }
+  if (exitCode === 2) return 'socket_error';
+  if (exitCode === 0 && stdout.trim().length > 0) return 'ok';
+  return 'no_reply';
+}
 
 export function pingCliAgent(timeoutMs = 30_000): Promise<PingResult> {
   return new Promise((resolve) => {
@@ -21,6 +35,7 @@ export function pingCliAgent(timeoutMs = 30_000): Promise<PingResult> {
       stdio: ['ignore', 'pipe', 'pipe'],
     });
     let stdout = '';
+    let stderr = '';
     let settled = false;
     const timer = setTimeout(() => {
       if (settled) return;
@@ -32,13 +47,14 @@ export function pingCliAgent(timeoutMs = 30_000): Promise<PingResult> {
     child.stdout.on('data', (chunk: Buffer) => {
       stdout += chunk.toString('utf-8');
     });
+    child.stderr.on('data', (chunk: Buffer) => {
+      stderr += chunk.toString('utf-8');
+    });
     child.on('close', (code) => {
       if (settled) return;
       settled = true;
       clearTimeout(timer);
-      if (code === 2) resolve('socket_error');
-      else if (code === 0 && stdout.trim().length > 0) resolve('ok');
-      else resolve('no_reply');
+      resolve(classifyPingResult(code, stdout, stderr));
     });
     child.on('error', () => {
       if (settled) return;

setup/lib/claude-assist.ts

@@ -64,6 +64,10 @@ const STEP_FILES: Record<string, string[]> = {
   'telegram-validate': ['setup/channels/telegram.ts'],
   'pair-telegram': ['setup/pair-telegram.ts', 'setup/channels/telegram.ts'],
   'discord-install': ['setup/add-discord.sh', 'setup/channels/discord.ts'],
+  'slack-install': ['setup/add-slack.sh', 'setup/channels/slack.ts'],
+  'slack-validate': ['setup/channels/slack.ts'],
+  'imessage-install': ['setup/add-imessage.sh', 'setup/channels/imessage.ts'],
+  'imessage': ['setup/channels/imessage.ts'],
   'teams-install': ['setup/add-teams.sh', 'setup/channels/teams.ts'],
   'teams-manifest': ['setup/lib/teams-manifest.ts', 'setup/channels/teams.ts'],
   'init-first-agent': [
@@ -115,7 +119,7 @@ export async function offerClaudeAssist(
   const run = ensureAnswer(
     await p.confirm({
       message: 'Run this command? (you can edit it before executing)',
-      initialValue: false,
+      initialValue: true,
     }),
   );
   if (!run) return false;
@@ -279,18 +283,24 @@ async function queryClaudeUnderSpinner(
     // No hard timeout — debugging can take a long time, and the cost of
     // cutting Claude off mid-investigation is worse than letting the
     // spinner run. The user can Ctrl-C if they want to abort.
-    const child = spawn(
-      'claude',
-      [
-        '-p',
-        '--output-format',
-        'stream-json',
-        '--verbose',
-        '--permission-mode',
-        'bypassPermissions',
-      ],
-      { cwd: projectRoot, stdio: ['pipe', 'pipe', 'pipe'] },
-    );
+    //
+    // Resume the same session on repeat invocations so Claude carries
+    // context across failures in one setup run.
+    const claudeArgs = [
+      '-p',
+      '--output-format',
+      'stream-json',
+      '--verbose',
+      '--permission-mode',
+      'bypassPermissions',
+    ];
+    if (claudeSessionId) {
+      claudeArgs.push('--resume', claudeSessionId);
+    }
+    const child = spawn('claude', claudeArgs, {
+      cwd: projectRoot,
+      stdio: ['pipe', 'pipe', 'pipe'],
+    });
 
     child.stdout.on('data', (c: Buffer) => {
       lineBuf += c.toString('utf-8');
@@ -301,6 +311,16 @@ async function queryClaudeUnderSpinner(
         if (!line.trim()) continue;
         try {
           const event = JSON.parse(line) as StreamEvent;
+          // Capture the session id on the very first claude invocation of
+          // this process so later calls can --resume it.
+          if (
+            !claudeSessionId &&
+            event.type === 'system' &&
+            event.subtype === 'init' &&
+            typeof event.session_id === 'string'
+          ) {
+            claudeSessionId = event.session_id;
+          }
           handleStreamEvent(event, {
             setAction: (a) => {
               actions.push(a);
@@ -335,10 +355,14 @@ async function queryClaudeUnderSpinner(
 }
 
 // Minimal shape of the stream-json events we care about. Claude emits
-// many more, but we only read tool_use blocks (for breadcrumbs) and text
-// blocks (to reassemble the final REASON/COMMAND answer).
+// many more, but we only read tool_use blocks (for breadcrumbs), text
+// blocks (to reassemble the final REASON/COMMAND answer), and the
+// session_id on the init event so follow-up invocations can resume the
+// same conversation.
 interface StreamEvent {
   type: string;
+  subtype?: string;
+  session_id?: string;
   message?: {
     content?: Array<
       | { type: 'text'; text: string }
@@ -347,6 +371,11 @@ interface StreamEvent {
   };
 }
 
+// The session id from the first claude-assist invocation in this process.
+// Subsequent invocations pass `--resume <id>` so Claude sees prior failures
+// as conversation history instead of treating each failure in isolation.
+let claudeSessionId: string | null = null;
+
 function handleStreamEvent(
   event: StreamEvent,
   cb: { setAction: (a: string) => void; appendText: (t: string) => void },

setup/lib/runner.ts

@@ -322,10 +322,12 @@ async function runUnderSpinner<
   if (result.ok) {
     const isSkipped = result.terminal?.fields.STATUS === 'skipped';
     const msg = isSkipped && labels.skipped ? labels.skipped : labels.done;
-    s.stop(`${fitToWidth(msg, suffix)}${k.dim(suffix)}`);
+    // Bold the outcome so the step's headline reads stronger than the prose
+    // body copy around it. The trailing `(Ns)` timing stays dim.
+    s.stop(`${k.bold(fitToWidth(msg, suffix))}${k.dim(suffix)}`);
   } else {
     const failMsg = labels.failed ?? labels.running.replace(/…$/, ' failed');
-    s.stop(`${fitToWidth(failMsg, suffix)}${k.dim(suffix)}`, 1);
+    s.stop(`${k.bold(fitToWidth(failMsg, suffix))}${k.dim(suffix)}`, 1);
     dumpTranscriptOnFailure(result.transcript);
   }
   return result;

setup/lib/setup-config-parse.ts

@@ -0,0 +1,161 @@
+/**
+ * Parser/reader/writer for the advanced-config registry (setup-config.ts).
+ *
+ *   readFromEnv()  → values found in process.env
+ *   parseFlags()   → values from argv, plus --help and any pass-through args
+ *   applyToEnv()   → write resolved values back to process.env so existing
+ *                    step code keeps reading env vars unchanged
+ *   printHelp()    → render --help from the registry
+ *
+ * Flag parsing supports:
+ *   --key value      space form
+ *   --key=value      equals form
+ *   --key            booleans only (sets true)
+ *   --no-key         booleans only (sets false)
+ */
+import {
+  CONFIG,
+  envVarFor,
+  flagFor,
+  findByFlag,
+  type Entry,
+} from './setup-config.js';
+
+export type ConfigValues = Record<string, string | boolean | number>;
+
+function coerce(e: Entry, raw: string): string | number | boolean | undefined {
+  switch (e.type) {
+    case 'boolean': {
+      const v = raw.toLowerCase();
+      if (['true', '1', 'yes'].includes(v)) return true;
+      if (['false', '0', 'no'].includes(v)) return false;
+      return undefined;
+    }
+    case 'integer': {
+      const n = Number(raw);
+      return Number.isFinite(n) ? n : undefined;
+    }
+    default:
+      return raw;
+  }
+}
+
+export function readFromEnv(env: NodeJS.ProcessEnv = process.env): ConfigValues {
+  const out: ConfigValues = {};
+  for (const e of CONFIG) {
+    const raw = env[envVarFor(e)];
+    if (raw === undefined || raw === '') continue;
+    const v = coerce(e, raw);
+    if (v !== undefined) out[e.key] = v;
+  }
+  return out;
+}
+
+export type FlagParseResult = {
+  values: ConfigValues;
+  rest: string[];
+  help: boolean;
+  errors: string[];
+};
+
+export function parseFlags(argv: string[]): FlagParseResult {
+  const values: ConfigValues = {};
+  const rest: string[] = [];
+  const errors: string[] = [];
+  let help = false;
+
+  for (let i = 0; i < argv.length; i++) {
+    const arg = argv[i];
+
+    if (arg === '--help' || arg === '-h') {
+      help = true;
+      continue;
+    }
+    // POSIX end-of-options. pnpm passes a bare `--` through when invoked as
+    // `pnpm run script --` with nothing after it; treat the rest as
+    // pass-through positional args.
+    if (arg === '--') {
+      rest.push(...argv.slice(i + 1));
+      break;
+    }
+    if (!arg.startsWith('--')) {
+      rest.push(arg);
+      continue;
+    }
+
+    const eq = arg.indexOf('=');
+    let name = eq === -1 ? arg : arg.slice(0, eq);
+    const inline: string | undefined = eq === -1 ? undefined : arg.slice(eq + 1);
+
+    let negated = false;
+    if (name.startsWith('--no-')) {
+      negated = true;
+      name = `--${name.slice(5)}`;
+    }
+
+    const entry = findByFlag(name);
+    if (!entry) {
+      errors.push(`Unknown flag: ${arg}`);
+      continue;
+    }
+
+    if (entry.type === 'boolean') {
+      if (negated) values[entry.key] = false;
+      else if (inline !== undefined) {
+        const v = coerce(entry, inline);
+        if (v === undefined) errors.push(`Invalid boolean for ${name}: ${inline}`);
+        else values[entry.key] = v;
+      } else values[entry.key] = true;
+      continue;
+    }
+
+    const raw = inline !== undefined ? inline : argv[++i];
+    if (raw === undefined) {
+      errors.push(`Missing value for ${name}`);
+      continue;
+    }
+    const v = coerce(entry, raw);
+    if (v === undefined) {
+      errors.push(`Invalid ${entry.type} for ${name}: ${raw}`);
+      continue;
+    }
+    if (entry.type === 'string' || entry.type === 'url') {
+      const err = entry.validate?.(raw);
+      if (err) {
+        errors.push(`${name}: ${err}`);
+        continue;
+      }
+    }
+    values[entry.key] = v;
+  }
+
+  return { values, rest, help, errors };
+}
+
+export function applyToEnv(
+  values: ConfigValues,
+  env: NodeJS.ProcessEnv = process.env,
+): void {
+  for (const e of CONFIG) {
+    if (!(e.key in values)) continue;
+    const v = values[e.key];
+    env[envVarFor(e)] =
+      typeof v === 'boolean' ? (v ? 'true' : 'false') : String(v);
+  }
+}
+
+export function printHelp(stream: NodeJS.WritableStream = process.stdout): void {
+  const lines: string[] = [];
+  lines.push('Usage: bash nanoclaw.sh [flags...]');
+  lines.push('');
+  lines.push('Flags:');
+  const width = Math.max(...CONFIG.map((e) => flagFor(e).length));
+  for (const e of CONFIG) {
+    const flag = flagFor(e).padEnd(width + 2);
+    lines.push(`  ${flag}${e.help}`);
+  }
+  lines.push('');
+  lines.push('Each flag also reads from its corresponding NANOCLAW_<KEY> env var.');
+  lines.push('Run without flags for the default interactive flow.');
+  stream.write(lines.join('\n') + '\n');
+}

setup/lib/setup-config-screen.ts

@@ -0,0 +1,127 @@
+/**
+ * Advanced-settings screen — menu of UI-visible entries from the config
+ * registry. The user picks one entry, edits it, returns to the menu, and
+ * exits via "Done". Returns a fresh values object; the caller passes it to
+ * applyToEnv() so downstream step code reads them via env vars.
+ *
+ * Per-entry edit contract:
+ *   - Blank input on text/password/integer = leave current value unchanged.
+ *   - Enums get a synthetic "leave unchanged" first option.
+ *   - Booleans use confirm with the current value as initialValue.
+ *   - Secret entries mask the current value as bullets in hints/labels.
+ */
+import * as p from '@clack/prompts';
+
+import { brightSelect } from './bright-select.js';
+import { ensureAnswer } from './runner.js';
+import { CONFIG, type Entry } from './setup-config.js';
+import type { ConfigValues } from './setup-config-parse.js';
+
+const SKIP_SENTINEL = '__leave_unchanged__';
+const DONE_SENTINEL = '__done__';
+const MASK = '••••••••';
+
+export async function runAdvancedScreen(
+  initial: ConfigValues,
+): Promise<ConfigValues> {
+  const result: ConfigValues = { ...initial };
+  const visible = CONFIG.filter((e) => e.surface === 'flag+ui');
+
+  while (true) {
+    const options = [
+      ...visible.map((e) => ({
+        value: e.key,
+        label: e.label,
+        hint: hintFor(e, result),
+      })),
+      { value: DONE_SENTINEL, label: 'Done — continue with setup' },
+    ];
+
+    const choice = ensureAnswer(
+      await brightSelect<string>({
+        message: 'Pick a setting to override',
+        options,
+        initialValue: DONE_SENTINEL,
+      }),
+    ) as string;
+
+    if (choice === DONE_SENTINEL) return result;
+    const entry = visible.find((e) => e.key === choice);
+    if (entry) await promptOne(entry, result);
+  }
+}
+
+function hintFor(e: Entry, values: ConfigValues): string {
+  const v = values[e.key];
+  if (v === undefined) return 'not set';
+  if (e.secret) return MASK;
+  return String(v);
+}
+
+async function promptOne(e: Entry, values: ConfigValues): Promise<void> {
+  if (e.type === 'boolean') {
+    const init =
+      typeof values[e.key] === 'boolean'
+        ? (values[e.key] as boolean)
+        : (e.default ?? false);
+    const ans = ensureAnswer(
+      await p.confirm({ message: e.label, initialValue: init }),
+    );
+    values[e.key] = ans as boolean;
+    return;
+  }
+
+  if (e.type === 'enum') {
+    const ans = ensureAnswer(
+      await brightSelect<string>({
+        message: e.label,
+        options: [
+          { value: SKIP_SENTINEL, label: 'Leave unchanged' },
+          ...e.options,
+        ],
+        initialValue: SKIP_SENTINEL,
+      }),
+    );
+    if (ans !== SKIP_SENTINEL) values[e.key] = ans as string;
+    return;
+  }
+
+  if (e.type === 'integer') {
+    const ans = ensureAnswer(
+      await p.text({
+        message: e.label,
+        placeholder: e.default !== undefined ? String(e.default) : undefined,
+        validate: (v) => {
+          const s = (v ?? '').trim();
+          if (!s) return undefined;
+          const n = Number(s);
+          if (!Number.isFinite(n)) return 'Must be a number';
+          if (e.min !== undefined && n < e.min) return `Must be ≥ ${e.min}`;
+          if (e.max !== undefined && n > e.max) return `Must be ≤ ${e.max}`;
+          return undefined;
+        },
+      }),
+    );
+    const trimmed = ((ans as string) ?? '').trim();
+    if (trimmed) values[e.key] = Number(trimmed);
+    return;
+  }
+
+  // string | url
+  const validate = (v: string | undefined): string | undefined => {
+    const s = (v ?? '').trim();
+    if (!s) return undefined;
+    return e.validate?.(s);
+  };
+  const ans = ensureAnswer(
+    e.secret
+      ? await p.password({ message: e.label, validate })
+      : await p.text({
+          message: e.label,
+          placeholder: e.placeholder ?? e.default,
+          validate,
+        }),
+  );
+  const trimmed = ((ans as string) ?? '').trim();
+  if (trimmed) values[e.key] = trimmed;
+}

setup/lib/setup-config.ts

@@ -0,0 +1,142 @@
+/**
+ * Setup-time advanced-config registry.
+ *
+ * One source of truth for: CLI flags, env-var names, the advanced-settings
+ * screen, and `--help` output. The flag parser, env reader, and UI screen
+ * all consume this list and write resolved values back to `process.env` so
+ * existing step code keeps reading env vars unchanged.
+ *
+ * Default name conventions (overridable per entry):
+ *   key 'fooBar' → envVar 'NANOCLAW_FOO_BAR' → flag '--foo-bar'
+ *
+ * Surface levels:
+ *   'flag'     — CLI flag + env var only (debug/internal knobs)
+ *   'flag+ui'  — also shown in the advanced-settings screen
+ */
+
+export type EntrySurface = 'flag' | 'flag+ui';
+
+interface BaseEntry {
+  /** Canonical camelCase key. */
+  key: string;
+  /** Override of the auto-derived NANOCLAW_<UPPER_SNAKE> env var. */
+  envVar?: string;
+  /** Override of the auto-derived --kebab-case flag. */
+  flag?: string;
+  label: string;
+  help: string;
+  surface: EntrySurface;
+  /** UI section header. Entries without a group land in 'Other'. */
+  group?: string;
+  /** Mask in UI, redact in logs. */
+  secret?: boolean;
+}
+
+interface StringEntry extends BaseEntry {
+  type: 'string' | 'url';
+  default?: string;
+  placeholder?: string;
+  validate?: (v: string) => string | undefined;
+}
+
+interface EnumEntry extends BaseEntry {
+  type: 'enum';
+  options: { value: string; label: string; hint?: string }[];
+  default?: string;
+}
+
+interface BoolEntry extends BaseEntry {
+  type: 'boolean';
+  default?: boolean;
+}
+
+interface IntEntry extends BaseEntry {
+  type: 'integer';
+  default?: number;
+  min?: number;
+  max?: number;
+}
+
+export type Entry = StringEntry | EnumEntry | BoolEntry | IntEntry;
+
+const httpUrl = (v: string): string | undefined =>
+  /^https?:\/\/\S+/.test(v) ? undefined : 'Must be http(s)://…';
+
+export const CONFIG: Entry[] = [
+  {
+    key: 'onecliApiHost',
+    label: 'OneCLI vault URL',
+    help: 'Use a remote OneCLI vault instead of installing one locally.',
+    surface: 'flag+ui',
+    group: 'OneCLI',
+    type: 'url',
+    default: 'https://app.onecli.sh',
+    placeholder: 'https://app.onecli.sh',
+    validate: httpUrl,
+  },
+  {
+    key: 'onecliApiToken',
+    label: 'OneCLI access token',
+    help: 'Bearer token for the remote vault. Required if --onecli-api-host is set.',
+    surface: 'flag+ui',
+    group: 'OneCLI',
+    type: 'string',
+    secret: true,
+    placeholder: 'oc_…',
+    validate: (v) => (v.startsWith('oc_') ? undefined : 'Must start with oc_'),
+  },
+  {
+    key: 'anthropicBaseUrl',
+    label: 'Anthropic API base URL',
+    help: 'Use a proxy or alternative endpoint instead of api.anthropic.com.',
+    surface: 'flag+ui',
+    group: 'Anthropic',
+    type: 'url',
+    placeholder: 'https://api.anthropic.com',
+    validate: httpUrl,
+  },
+  {
+    key: 'anthropicAuthToken',
+    label: 'Anthropic auth token',
+    help: 'Bearer token for the custom Anthropic endpoint. Used together with --anthropic-base-url.',
+    surface: 'flag+ui',
+    group: 'Anthropic',
+    type: 'string',
+    secret: true,
+    validate: (v) => (v.trim() ? undefined : 'Required'),
+  },
+
+  // Existing env-var knobs — flag-only so they don't clutter the UI screen.
+  {
+    key: 'skip',
+    envVar: 'NANOCLAW_SKIP',
+    label: 'Skip steps',
+    help: 'Comma-separated step names to skip (debugging only).',
+    surface: 'flag',
+    type: 'string',
+  },
+  {
+    key: 'displayName',
+    envVar: 'NANOCLAW_DISPLAY_NAME',
+    label: 'Display name',
+    help: 'Skip the "what should your assistant call you?" prompt.',
+    surface: 'flag',
+    type: 'string',
+  },
+];
+
+// ─── name derivation ───────────────────────────────────────────────────
+
+export function envVarFor(e: Entry): string {
+  if (e.envVar) return e.envVar;
+  return `NANOCLAW_${e.key.replace(/[A-Z]/g, (c) => `_${c}`).toUpperCase()}`;
+}
+
+export function flagFor(e: Entry): string {
+  if (e.flag) return e.flag;
+  return `--${e.key.replace(/[A-Z]/g, (c) => `-${c.toLowerCase()}`)}`;
+}
+
+export function findByFlag(flag: string): Entry | null {
+  return CONFIG.find((e) => flagFor(e) === flag) ?? null;
+}

setup/lib/theme.ts

@@ -58,17 +58,14 @@ export function wrapForGutter(text: string, gutter: number): string {
 }
 
 /**
- * Wrap + dim together. Needed instead of `k.dim(wrapForGutter(...))`
- * because clack resets styling at each line break when rendering
- * multi-line log content — a single outer dim envelope only colors the
- * first line. Applying dim per-line gives each wrapped row its own
- * `\x1b[2m…\x1b[0m` envelope so the whole block reads as one block.
+ * Wrap multi-line explanatory prose to the clack gutter. Previously
+ * dimmed its output (hence the name) — that made body copy hard to read
+ * against dark terminals. Dim is now reserved for preview/debug blocks
+ * (failure transcript tails, claude-assist streams); prose renders at
+ * the terminal's regular weight.
  */
 export function dimWrap(text: string, gutter: number): string {
-  return wrapForGutter(text, gutter)
-    .split('\n')
-    .map((line) => k.dim(line))
-    .join('\n');
+  return wrapForGutter(text, gutter);
 }
 
 const ANSI_RE = /\x1b\[[0-9;]*m/g;

setup/onecli.ts

@@ -86,40 +86,161 @@ function ensureShellProfilePath(): void {
   }
 }
 
-function writeEnvOnecliUrl(url: string): void {
+function writeEnvVar(name: string, value: string): void {
   const envFile = path.join(process.cwd(), '.env');
   let content = fs.existsSync(envFile) ? fs.readFileSync(envFile, 'utf-8') : '';
-  if (/^ONECLI_URL=/m.test(content)) {
-    content = content.replace(/^ONECLI_URL=.*$/m, `ONECLI_URL=${url}`);
+  const re = new RegExp(`^${name}=.*$`, 'm');
+  if (re.test(content)) {
+    content = content.replace(re, `${name}=${value}`);
   } else {
-    content = content.trimEnd() + (content ? '\n' : '') + `ONECLI_URL=${url}\n`;
+    content = content.trimEnd() + (content ? '\n' : '') + `${name}=${value}\n`;
   }
   fs.writeFileSync(envFile, content);
 }
 
+function writeEnvOnecliUrl(url: string): void {
+  writeEnvVar('ONECLI_URL', url);
+}
+
+// Last-known-good CLI release. Used only if BOTH the upstream installer
+// and the redirect-based version probe fail. Bump deliberately when a
+// new CLI release ships.
+const ONECLI_CLI_FALLBACK_VERSION = '1.3.0';
+const ONECLI_CLI_REPO = 'onecli/onecli-cli';
+
+function installOnecliCliOnly(): { stdout: string; ok: boolean } {
+  const upstream = runInstall('curl -fsSL onecli.sh/cli/install | sh');
+  if (upstream.ok) return { stdout: upstream.stdout, ok: true };
+  const fallback = installOnecliCliDirect();
+  return { stdout: upstream.stdout + (upstream.stderr ?? '') + '\n' + fallback.stdout, ok: fallback.ok };
+}
+
 function installOnecli(): { stdout: string; ok: boolean } {
-  // OneCLI's own install script handles gateway + CLI + PATH.
-  // We run the two canonical installers in sequence and capture stdout so
-  // we can extract the printed URL as a fallback to `onecli config get`.
   let stdout = '';
+
+  // Gateway install (docker-compose based, no rate-limit concerns).
+  const gw = runInstall('curl -fsSL onecli.sh/install | sh');
+  stdout += gw.stdout;
+  if (!gw.ok) {
+    log.error('OneCLI gateway install failed', { stderr: gw.stderr });
+    return { stdout: stdout + (gw.stderr ?? ''), ok: false };
+  }
+
+  // CLI install. The upstream script calls the GitHub releases API
+  // (api.github.com) to resolve the latest tag — which 403s anonymous
+  // callers after 60 requests/hour per IP. Try upstream first; on failure
+  // resolve the version ourselves (via HTTP redirect, which isn't
+  // API-throttled) and download the release archive directly.
+  const upstream = runInstall('curl -fsSL onecli.sh/cli/install | sh');
+  stdout += upstream.stdout;
+  if (upstream.ok) return { stdout, ok: true };
+
+  log.warn('Upstream CLI installer failed — falling back to direct download', {
+    stderr: upstream.stderr,
+  });
+  stdout += (upstream.stderr ?? '') + '\n';
+
+  const fallback = installOnecliCliDirect();
+  stdout += fallback.stdout;
+  if (!fallback.ok) {
+    log.error('OneCLI CLI install failed (both upstream and direct fallback)');
+    return { stdout, ok: false };
+  }
+  return { stdout, ok: true };
+}
+
+function runInstall(cmd: string): { stdout: string; stderr?: string; ok: boolean } {
   try {
-    stdout += execSync('curl -fsSL onecli.sh/install | sh', {
-      encoding: 'utf-8',
-      stdio: ['ignore', 'pipe', 'pipe'],
-    });
-    stdout += execSync('curl -fsSL onecli.sh/cli/install | sh', {
+    const stdout = execSync(cmd, {
       encoding: 'utf-8',
       stdio: ['ignore', 'pipe', 'pipe'],
     });
     return { stdout, ok: true };
   } catch (err) {
     const e = err as { stdout?: string; stderr?: string };
-    log.error('OneCLI install failed', { stderr: e.stderr });
-    return { stdout: stdout + (e.stdout ?? '') + (e.stderr ?? ''), ok: false };
+    return { stdout: e.stdout ?? '', stderr: e.stderr, ok: false };
   }
 }
 
-async function pollHealth(url: string, timeoutMs: number): Promise<boolean> {
+/**
+ * Reinstate the OneCLI CLI install without hitting GitHub's rate-limited
+ * releases API. Resolves the version via the HTTP redirect from
+ * /releases/latest → /releases/tag/vX.Y.Z, then downloads the archive
+ * directly. Falls back to ONECLI_CLI_FALLBACK_VERSION if the redirect
+ * probe also fails.
+ */
+function installOnecliCliDirect(): { stdout: string; ok: boolean } {
+  const lines: string[] = [];
+  const append = (s: string): void => {
+    lines.push(s);
+  };
+
+  const osName = process.platform === 'darwin' ? 'darwin' : process.platform === 'linux' ? 'linux' : null;
+  if (!osName) {
+    append(`Unsupported platform: ${process.platform}`);
+    return { stdout: lines.join('\n'), ok: false };
+  }
+  const arch = process.arch === 'x64' ? 'amd64' : process.arch === 'arm64' ? 'arm64' : null;
+  if (!arch) {
+    append(`Unsupported arch: ${process.arch}`);
+    return { stdout: lines.join('\n'), ok: false };
+  }
+
+  let version: string | null = null;
+  try {
+    const redirect = execSync(
+      `curl -fsSL -o /dev/null -w '%{url_effective}' https://github.com/${ONECLI_CLI_REPO}/releases/latest`,
+      { encoding: 'utf-8', stdio: ['ignore', 'pipe', 'pipe'] },
+    ).trim();
+    const m = redirect.match(/\/tag\/v?([^/]+)$/);
+    if (m) version = m[1];
+  } catch {
+    // redirect probe failed — we'll pin the fallback
+  }
+  if (!version) {
+    version = ONECLI_CLI_FALLBACK_VERSION;
+    append(`Version probe failed; installing pinned fallback ${version}.`);
+  } else {
+    append(`Resolved onecli CLI ${version} via release redirect.`);
+  }
+
+  const archive = `onecli_${version}_${osName}_${arch}.tar.gz`;
+  const url = `https://github.com/${ONECLI_CLI_REPO}/releases/download/v${version}/${archive}`;
+  const tmpDir = fs.mkdtempSync(path.join(os.tmpdir(), 'onecli-'));
+  const archivePath = path.join(tmpDir, archive);
+
+  try {
+    append(`Downloading ${url}`);
+    execSync(`curl -fsSL -o ${JSON.stringify(archivePath)} ${JSON.stringify(url)}`, {
+      stdio: ['ignore', 'pipe', 'pipe'],
+    });
+    execSync(`tar -xzf ${JSON.stringify(archivePath)} -C ${JSON.stringify(tmpDir)}`, {
+      stdio: ['ignore', 'pipe', 'pipe'],
+    });
+
+    let installDir = '/usr/local/bin';
+    try {
+      fs.accessSync(installDir, fs.constants.W_OK);
+    } catch {
+      installDir = LOCAL_BIN;
+      fs.mkdirSync(installDir, { recursive: true });
+    }
+    const binSrc = path.join(tmpDir, 'onecli');
+    const binDest = path.join(installDir, 'onecli');
+    fs.copyFileSync(binSrc, binDest);
+    fs.chmodSync(binDest, 0o755);
+    append(`onecli ${version} installed to ${binDest}.`);
+    return { stdout: lines.join('\n'), ok: true };
+  } catch (err) {
+    const e = err as { stdout?: string; stderr?: string; message?: string };
+    append(`Direct install failed: ${e.stderr ?? e.message ?? String(err)}`);
+    return { stdout: lines.join('\n'), ok: false };
+  } finally {
+    fs.rmSync(tmpDir, { recursive: true, force: true });
+  }
+}
+
+export async function pollHealth(url: string, timeoutMs: number): Promise<boolean> {
   // `/api/health` matches the path probe.sh uses — keep them aligned.
   const deadline = Date.now() + timeoutMs;
   while (Date.now() < deadline) {
@@ -136,8 +257,64 @@ async function pollHealth(url: string, timeoutMs: number): Promise<boolean> {
 
 export async function run(args: string[]): Promise<void> {
   const reuse = args.includes('--reuse');
+  const remoteUrlIdx = args.indexOf('--remote-url');
+  const remoteUrl = remoteUrlIdx !== -1 ? args[remoteUrlIdx + 1] : null;
   ensureShellProfilePath();
 
+  if (remoteUrl) {
+    // Remote-mode: install only the CLI, point it at the remote gateway, and
+    // record the URL in .env. No local gateway is started.
+    log.info('Installing OneCLI CLI for remote gateway', { remoteUrl });
+    const res = installOnecliCliOnly();
+    if (!res.ok || !onecliVersion()) {
+      emitStatus('ONECLI', {
+        INSTALLED: false,
+        STATUS: 'failed',
+        ERROR: 'cli_install_failed',
+        HINT: 'CLI binary install failed. Make sure curl is installed and ~/.local/bin is writable.',
+        LOG: 'logs/setup.log',
+      });
+      process.exit(1);
+    }
+    try {
+      execFileSync('onecli', ['config', 'set', 'api-host', remoteUrl], {
+        stdio: 'ignore',
+        env: childEnv(),
+      });
+    } catch (err) {
+      log.warn('onecli config set api-host failed', { err });
+    }
+    writeEnvOnecliUrl(remoteUrl);
+    log.info('Wrote ONECLI_URL to .env', { url: remoteUrl });
+    const remoteToken = process.env.NANOCLAW_ONECLI_API_TOKEN?.trim();
+    if (remoteToken) {
+      // Two auth surfaces: `onecli auth login` persists the key for CLI
+      // calls during setup itself (e.g. detecting an existing Anthropic
+      // secret via `onecli secrets list`), and ONECLI_API_KEY in .env is
+      // read by the runtime SDK at request time. Both are needed.
+      try {
+        execFileSync('onecli', ['auth', 'login', '--api-key', remoteToken], {
+          stdio: 'ignore',
+          env: childEnv(),
+        });
+      } catch (err) {
+        log.warn('onecli auth login failed', { err });
+      }
+      writeEnvVar('ONECLI_API_KEY', remoteToken);
+      log.info('Wrote ONECLI_API_KEY to .env');
+    }
+    const healthy = await pollHealth(remoteUrl, 5000);
+    emitStatus('ONECLI', {
+      INSTALLED: true,
+      REMOTE: true,
+      ONECLI_URL: remoteUrl,
+      HEALTHY: healthy,
+      STATUS: 'success',
+      LOG: 'logs/setup.log',
+    });
+    return;
+  }
+
   if (reuse) {
     // Reuse-mode: don't touch the running gateway at all. Just verify it
     // exists, read its api-host, write ONECLI_URL to .env, and move on.

setup/peer-cleanup.ts

@@ -0,0 +1,186 @@
+/**
+ * Detect and clean up unhealthy NanoClaw peer services.
+ *
+ * Runs as a setup preflight before we install our own service. A crash-looping
+ * peer install (typically the legacy v1 `com.nanoclaw` plist) silently trashes
+ * this install's containers on every respawn because its `cleanupOrphans()`
+ * reaps anything matching `nanoclaw-`. We scope our reaper by label now, but
+ * we still need to stop the peer from killing us on its way down.
+ *
+ * A peer is "unhealthy" when:
+ *   - launchd: `state != running` AND `runs > UNHEALTHY_RUNS_THRESHOLD`
+ *   - systemd: unit is in `failed` state, OR `activating` with many restarts
+ *
+ * Healthy peers are left alone — multiple installs can coexist fine now that
+ * container-reaper is label-scoped.
+ */
+import { execFileSync } from 'child_process';
+import fs from 'fs';
+import os from 'os';
+import path from 'path';
+
+import { getLaunchdLabel, getSystemdUnit } from '../src/install-slug.js';
+import { log } from '../src/log.js';
+
+const UNHEALTHY_RUNS_THRESHOLD = 10;
+
+export interface PeerStatus {
+  label: string;
+  configPath: string;
+  state: string;
+  runs: number;
+  unhealthy: boolean;
+}
+
+export interface PeerCleanupResult {
+  checked: PeerStatus[];
+  unloaded: PeerStatus[];
+  failures: Array<{ label: string; err: string }>;
+}
+
+/**
+ * Scan for peer NanoClaw services and unload any that are crash-looping.
+ * Returns a summary suitable for emitStatus / setup-log reporting.
+ */
+export function cleanupUnhealthyPeers(projectRoot: string = process.cwd()): PeerCleanupResult {
+  const platform = os.platform();
+  if (platform === 'darwin') {
+    return cleanupLaunchdPeers(projectRoot);
+  }
+  if (platform === 'linux') {
+    return cleanupSystemdPeers(projectRoot);
+  }
+  return { checked: [], unloaded: [], failures: [] };
+}
+
+// ---- launchd (macOS) --------------------------------------------------------
+
+function cleanupLaunchdPeers(projectRoot: string): PeerCleanupResult {
+  const ownLabel = getLaunchdLabel(projectRoot);
+  const agentsDir = path.join(os.homedir(), 'Library', 'LaunchAgents');
+  const result: PeerCleanupResult = { checked: [], unloaded: [], failures: [] };
+
+  let plists: string[];
+  try {
+    plists = fs
+      .readdirSync(agentsDir)
+      .filter((f) => /^com\.nanoclaw.*\.plist$/.test(f))
+      .map((f) => path.join(agentsDir, f));
+  } catch {
+    return result;
+  }
+
+  const uid = process.getuid?.() ?? 0;
+
+  for (const plistPath of plists) {
+    const label = path.basename(plistPath, '.plist');
+    if (label === ownLabel) continue;
+
+    const status = probeLaunchdPeer(label, plistPath, uid);
+    if (!status) continue;
+    result.checked.push(status);
+
+    if (!status.unhealthy) continue;
+
+    try {
+      execFileSync('launchctl', ['unload', plistPath], { stdio: 'pipe' });
+      log.info('Unloaded unhealthy peer launchd service', {
+        label,
+        state: status.state,
+        runs: status.runs,
+        plistPath,
+      });
+      result.unloaded.push(status);
+    } catch (err) {
+      const message = err instanceof Error ? err.message : String(err);
+      log.warn('Failed to unload peer launchd service', { label, err: message });
+      result.failures.push({ label, err: message });
+    }
+  }
+
+  return result;
+}
+
+function probeLaunchdPeer(label: string, plistPath: string, uid: number): PeerStatus | null {
+  let output: string;
+  try {
+    output = execFileSync('launchctl', ['print', `gui/${uid}/${label}`], {
+      stdio: ['ignore', 'pipe', 'pipe'],
+      encoding: 'utf-8',
+    });
+  } catch {
+    // Not loaded → not currently a threat. Skip silently.
+    return null;
+  }
+
+  const state = /^\s*state\s*=\s*(.+?)\s*$/m.exec(output)?.[1] ?? 'unknown';
+  const runsStr = /^\s*runs\s*=\s*(\d+)/m.exec(output)?.[1];
+  const runs = runsStr ? parseInt(runsStr, 10) : 0;
+
+  const unhealthy = state !== 'running' && runs > UNHEALTHY_RUNS_THRESHOLD;
+  return { label, configPath: plistPath, state, runs, unhealthy };
+}
+
+// ---- systemd (Linux) --------------------------------------------------------
+
+function cleanupSystemdPeers(projectRoot: string): PeerCleanupResult {
+  const ownUnit = getSystemdUnit(projectRoot);
+  const unitDir = path.join(os.homedir(), '.config', 'systemd', 'user');
+  const result: PeerCleanupResult = { checked: [], unloaded: [], failures: [] };
+
+  let units: string[];
+  try {
+    units = fs
+      .readdirSync(unitDir)
+      .filter((f) => /^nanoclaw.*\.service$/.test(f))
+      .map((f) => f.replace(/\.service$/, ''));
+  } catch {
+    return result;
+  }
+
+  for (const unit of units) {
+    if (unit === ownUnit) continue;
+
+    const status = probeSystemdPeer(unit);
+    if (!status) continue;
+    result.checked.push(status);
+
+    if (!status.unhealthy) continue;
+
+    try {
+      execFileSync('systemctl', ['--user', 'disable', '--now', `${unit}.service`], { stdio: 'pipe' });
+      log.info('Disabled unhealthy peer systemd unit', {
+        unit,
+        state: status.state,
+        runs: status.runs,
+      });
+      result.unloaded.push(status);
+    } catch (err) {
+      const message = err instanceof Error ? err.message : String(err);
+      log.warn('Failed to disable peer systemd unit', { unit, err: message });
+      result.failures.push({ label: unit, err: message });
+    }
+  }
+
+  return result;
+}
+
+function probeSystemdPeer(unit: string): PeerStatus | null {
+  const unitPath = path.join(os.homedir(), '.config', 'systemd', 'user', `${unit}.service`);
+  try {
+    const output = execFileSync(
+      'systemctl',
+      ['--user', 'show', '--property=ActiveState,NRestarts', `${unit}.service`],
+      { stdio: ['ignore', 'pipe', 'pipe'], encoding: 'utf-8' },
+    );
+    const activeState = /^ActiveState=(.+)$/m.exec(output)?.[1]?.trim() ?? 'unknown';
+    const restartsStr = /^NRestarts=(\d+)/m.exec(output)?.[1];
+    const runs = restartsStr ? parseInt(restartsStr, 10) : 0;
+
+    const unhealthy =
+      activeState === 'failed' || (activeState !== 'active' && runs > UNHEALTHY_RUNS_THRESHOLD);
+    return { label: unit, configPath: unitPath, state: activeState, runs, unhealthy };
+  } catch {
+    return null;
+  }
+}

setup/providers/codex-registration.test.ts

@@ -0,0 +1,22 @@
+/**
+ * Setup-side registration guard for the codex provider (the third barrel of
+ * the multi-point archetype): imports the REAL setup/providers barrel and
+ * asserts the registry carries codex with its auth + install check. Red if
+ * the barrel line is deleted, the barrel fails to evaluate, or the payload
+ * module breaks. (Importing ./codex.js directly would self-register and stay
+ * green when the barrel line is deleted.)
+ */
+import { describe, expect, it } from 'vitest';
+
+import { getSetupProvider } from './registry.js';
+import './index.js'; // the real setup provider barrel
+
+describe('codex setup registration', () => {
+  it('registers codex with auth + install check via the barrel', () => {
+    const codex = getSetupProvider('codex');
+    expect(codex).toBeDefined();
+    expect(typeof codex!.runAuth).toBe('function');
+    expect(typeof codex!.runInstallCheck).toBe('function');
+    expect(typeof codex!.offerFailureAssist).toBe('function');
+  });
+});

setup/providers/codex.test.ts

@@ -0,0 +1,101 @@
+import { EventEmitter } from 'events';
+import fs from 'fs';
+import os from 'os';
+import path from 'path';
+
+import { describe, expect, it, vi } from 'vitest';
+
+// Mock child_process so runCodexLoginAuth never spawns a real codex CLI; the
+// spawn stand-in plays `codex login` writing auth.json into whatever
+// CODEX_HOME it was handed.
+const mockSpawn = vi.fn();
+const mockSpawnSync = vi.fn();
+const mockExecFileSync = vi.fn();
+vi.mock('child_process', () => ({
+  spawn: (...args: unknown[]) => mockSpawn(...args),
+  spawnSync: (...args: unknown[]) => mockSpawnSync(...args),
+  execFileSync: (...args: unknown[]) => mockExecFileSync(...args),
+}));
+
+// Keep the auth flow's structured logging out of logs/setup.log.
+vi.mock('../logs.js', () => ({ step: vi.fn(), userInput: vi.fn() }));
+
+import { buildCodexFailurePrompt, runCodexLoginAuth, verifyCodexInstall } from './codex.js';
+
+// Structural guard for the codex payload wiring: provider files, both barrel
+// imports, and the pinned Dockerfile install. Goes red if any of them is
+// removed without going through the /add-codex (or its REMOVE.md) path.
+describe('verifyCodexInstall', () => {
+  it('passes on a tree with the codex payload wired', () => {
+    const { ok, problems } = verifyCodexInstall();
+    expect(problems).toEqual([]);
+    expect(ok).toBe(true);
+  });
+});
+
+// Pure prompt builder for the failure-assist hook — no spawning involved.
+describe('buildCodexFailurePrompt', () => {
+  it('carries the failure context and the de-duped reference list', () => {
+    const projectRoot = '/repo';
+    const prompt = buildCodexFailurePrompt(
+      {
+        stepName: 'verify',
+        msg: 'first-chat ping timed out',
+        hint: 'check the container logs',
+        rawLogPath: '/repo/logs/setup-steps/verify.log',
+      },
+      projectRoot,
+    );
+
+    expect(prompt).toContain('Failed step: verify');
+    expect(prompt).toContain('Error: first-chat ping timed out');
+    expect(prompt).toContain('Hint: check the container logs');
+    expect(prompt).toContain('README.md'); // BIG_PICTURE_FILES
+    expect(prompt).toContain('setup/verify.ts'); // STEP_FILES['verify']
+    expect(prompt).toContain('logs/setup.log');
+    expect(prompt).toContain('logs/setup-steps/verify.log'); // relativized rawLogPath
+  });
+
+  it('falls back to the step-log directory when no raw log path is given', () => {
+    const prompt = buildCodexFailurePrompt({ stepName: 'verify', msg: 'boom' }, '/repo');
+    expect(prompt).toContain('logs/setup-steps/');
+    expect(prompt).not.toContain('Hint:');
+  });
+});
+
+// Session-isolation invariant: the ChatGPT session vaulted for the gateway
+// must never be the user's personal ~/.codex session — sharing one OAuth
+// session across two consumers gets the whole family invalidated server-side
+// when refresh tokens rotate (see the header of codex.ts).
+describe('runCodexLoginAuth', () => {
+  it('logs in under an isolated CODEX_HOME, vaults from it, and deletes it', async () => {
+    mockSpawnSync.mockReturnValue({ status: 0, stdout: '', stderr: '' });
+    mockExecFileSync.mockReturnValue('');
+
+    let loginEnv: NodeJS.ProcessEnv | undefined;
+    mockSpawn.mockImplementation((...args: unknown[]) => {
+      const opts = args[2] as { env?: NodeJS.ProcessEnv };
+      loginEnv = opts.env;
+      fs.writeFileSync(path.join(opts.env!.CODEX_HOME!, 'auth.json'), '{"tokens":{}}');
+      const child = new EventEmitter();
+      setImmediate(() => child.emit('close', 0));
+      return child;
+    });
+
+    await runCodexLoginAuth('browser');
+
+    // The login spawn ran under a CODEX_HOME that is not the personal one.
+    const codexHome = loginEnv?.CODEX_HOME;
+    expect(codexHome).toBeDefined();
+    expect(codexHome).not.toBe(path.join(os.homedir(), '.codex'));
+
+    // The vault snapshot was read from the isolated dir, not ~/.codex.
+    const vaultCall = mockExecFileSync.mock.calls.find((c) => c[0] === 'onecli');
+    expect(vaultCall).toBeDefined();
+    const vaultArgs = vaultCall![1] as string[];
+    expect(vaultArgs[vaultArgs.indexOf('--file') + 1]).toBe(path.join(codexHome!, 'auth.json'));
+
+    // The isolated dir holds a live credential — gone once vaulted.
+    expect(fs.existsSync(codexHome!)).toBe(false);
+  });
+});

setup/providers/codex.ts

@@ -0,0 +1,449 @@
+/**
+ * Codex provider setup — auth walk-through + install verification.
+ *
+ * Codex-owned payload code: when the codex provider moves to the `providers`
+ * branch, this file travels with it and `/add-codex` copies it back in. The
+ * only trunk reach-in is one import + one picker entry in setup/auto.ts.
+ *
+ * Auth honors the v2 credential invariant — everything lands in the OneCLI
+ * vault, nothing in .env, nothing in the container:
+ *   - ChatGPT subscription (the common case): `codex login` (browser) or
+ *     `codex login --device-auth` (URL + pairing code) runs with CODEX_HOME
+ *     pointed at a throwaway dir; the auth.json written there is stored
+ *     WHOLE in the vault (`--file … --host-pattern chatgpt.com`) and the dir
+ *     is deleted. The gateway injects it in flight; the container only ever
+ *     sees the `onecli-managed` placeholder.
+ *   - API key: pasted once, stored as an `openai` secret for api.openai.com.
+ *
+ * Session-isolation invariant: the vaulted ChatGPT session must be DEDICATED
+ * to the gateway. Never vault a copy of the user's live ~/.codex/auth.json.
+ * OpenAI rotates refresh tokens, so two consumers sharing one OAuth session
+ * strand each other on refresh, and replaying the stale token trips reuse
+ * detection — which invalidates the whole session family server-side
+ * (`token_invalidated`) for the gateway AND the user's personal Codex CLI.
+ */
+import { execFileSync, spawn, spawnSync } from 'child_process';
+import fs from 'fs';
+import os from 'os';
+import path from 'path';
+
+import * as p from '@clack/prompts';
+import k from 'kleur';
+
+import { brightSelect } from '../lib/bright-select.js';
+import { type AssistContext, BIG_PICTURE_FILES, STEP_FILES } from '../lib/claude-assist.js';
+import { brandBody, note } from '../lib/theme.js';
+import * as setupLog from '../logs.js';
+import { type FailureAssistResult, registerSetupProvider } from './registry.js';
+
+// ─── OneCLI vault helpers ────────────────────────────────────────────────
+
+interface OnecliSecret {
+  id: string;
+  name: string;
+  type: string;
+  hostPattern: string | null;
+}
+
+function listSecrets(): OnecliSecret[] {
+  const out = execFileSync('onecli', ['secrets', 'list'], { encoding: 'utf-8' });
+  const parsed = JSON.parse(out) as { data?: unknown };
+  return Array.isArray(parsed.data) ? (parsed.data as OnecliSecret[]) : [];
+}
+
+function findOpenAISecret(secrets: OnecliSecret[]): OnecliSecret | undefined {
+  return secrets.find((s) => {
+    const name = s.name.toLowerCase();
+    const type = s.type.toLowerCase();
+    const hostPattern = (s.hostPattern ?? '').toLowerCase();
+    return (
+      name === 'codex' ||
+      name === 'openai' ||
+      type === 'openai' ||
+      hostPattern.includes('api.openai.com') ||
+      hostPattern.includes('chatgpt.com')
+    );
+  });
+}
+
+function openAISecretExists(): boolean {
+  try {
+    return findOpenAISecret(listSecrets()) !== undefined;
+  } catch {
+    return false;
+  }
+}
+
+// ─── auth step ───────────────────────────────────────────────────────────
+
+function ensureAnswer<T>(value: T | symbol): T {
+  if (p.isCancel(value)) {
+    p.cancel('Setup cancelled.');
+    process.exit(1);
+  }
+  return value as T;
+}
+
+export async function runCodexAuthStep(): Promise<void> {
+  if (openAISecretExists()) {
+    p.log.success(brandBody('Your OpenAI account is already connected.'));
+    setupLog.step('auth', 'skipped', 0, { REASON: 'openai-secret-already-present', PROVIDER: 'codex' });
+    return;
+  }
+
+  const method = ensureAnswer(
+    await brightSelect<'browser' | 'device' | 'api' | 'skip'>({
+      message: 'How would you like to connect Codex?',
+      options: [
+        {
+          value: 'browser',
+          label: 'Sign in with my ChatGPT subscription',
+          hint: 'recommended if you have Plus or Pro — opens a browser',
+        },
+        {
+          value: 'device',
+          label: 'ChatGPT device pairing',
+          hint: 'no browser handoff — shows a URL and a code',
+        },
+        {
+          value: 'api',
+          label: 'Paste an OpenAI API key',
+          hint: 'pay-per-use; stored in OneCLI, never copied into the container',
+        },
+        {
+          value: 'skip',
+          label: "Skip — I'll connect later",
+          hint: 'Codex groups will start, but model calls will fail auth',
+        },
+      ],
+    }),
+  );
+  setupLog.userInput('codex_auth_method', method);
+
+  if (method === 'skip') {
+    const confirmed = ensureAnswer(
+      await p.confirm({
+        message: "Skip Codex sign-in? Codex won't be able to answer until you connect an OpenAI account.",
+        initialValue: false,
+      }),
+    );
+    if (!confirmed) return runCodexAuthStep();
+    setupLog.step('auth', 'skipped', 0, { REASON: 'user-skipped', PROVIDER: 'codex' });
+    p.log.warn(brandBody('Codex sign-in skipped. Add an OpenAI account to OneCLI before using Codex groups.'));
+    return;
+  }
+
+  if (method === 'api') {
+    await runCodexApiKeyAuth();
+    return;
+  }
+
+  await runCodexLoginAuth(method);
+}
+
+async function runCodexApiKeyAuth(): Promise<void> {
+  const key = ensureAnswer(
+    await p.password({
+      message: 'Paste your OpenAI API key (sk-…)',
+      validate: (v) => (v && v.trim().startsWith('sk-') ? undefined : 'That does not look like an OpenAI API key.'),
+    }),
+  ) as string;
+
+  try {
+    execFileSync(
+      'onecli',
+      [
+        'secrets',
+        'create',
+        '--name',
+        'Codex',
+        '--type',
+        'openai',
+        '--value',
+        key.trim(),
+        '--host-pattern',
+        'api.openai.com',
+      ],
+      { stdio: ['ignore', 'pipe', 'pipe'] },
+    );
+  } catch (err) {
+    setupLog.step('auth', 'failed', 0, { PROVIDER: 'codex', METHOD: 'api', ERROR: String(err) });
+    p.log.error(
+      brandBody(
+        "Couldn't save your OpenAI key to the vault. Make sure OneCLI is running (`onecli version`), then retry.",
+      ),
+    );
+    process.exit(1);
+  }
+  setupLog.step('auth', 'success', 0, { PROVIDER: 'codex', METHOD: 'api' });
+  p.log.success(brandBody('OpenAI account connected.'));
+}
+
+export async function runCodexLoginAuth(method: 'browser' | 'device'): Promise<void> {
+  const codexCheck = spawnSync('codex', ['--version'], { encoding: 'utf-8', stdio: ['ignore', 'pipe', 'pipe'] });
+  if (codexCheck.status !== 0) {
+    p.log.error(
+      brandBody(
+        'The Codex CLI is not installed on this machine. Install it with `npm install -g @openai/codex`, then re-run setup — or choose the API key option instead.',
+      ),
+    );
+    setupLog.step('auth', 'failed', 0, { PROVIDER: 'codex', METHOD: method, ERROR: 'codex_cli_missing' });
+    process.exit(1);
+  }
+
+  if (method === 'browser') {
+    p.log.step(brandBody('Opening the Codex sign-in flow…'));
+    console.log(k.dim('   (a browser will open for sign-in; this part is interactive)'));
+  } else {
+    p.log.step(brandBody('Starting Codex device-code pairing…'));
+    console.log(k.dim('   (a URL and code will appear below — open the URL and enter the code)'));
+  }
+  console.log();
+
+  // Session-isolation invariant (see file header): the login runs under a
+  // throwaway CODEX_HOME so the vaulted session is dedicated to the gateway
+  // and never shared with the user's personal ~/.codex.
+  const loginHome = fs.mkdtempSync(path.join(os.tmpdir(), 'codex-vault-login-'));
+  // Holds a live credential after login — must go on every exit path. The
+  // failure branches call process.exit, which skips finally blocks, so each
+  // removes it explicitly.
+  const removeLoginHome = (): void => fs.rmSync(loginHome, { recursive: true, force: true });
+
+  const args = method === 'device' ? ['login', '--device-auth'] : ['login'];
+  const start = Date.now();
+  const code = await runInherit('codex', args, { CODEX_HOME: loginHome });
+  const durationMs = Date.now() - start;
+  console.log();
+
+  if (code !== 0) {
+    removeLoginHome();
+    setupLog.step('auth', 'failed', durationMs, { PROVIDER: 'codex', METHOD: method, EXIT_CODE: String(code) });
+    p.log.error(
+      brandBody(
+        "Couldn't complete the Codex sign-in. Re-run setup and try again, or choose the API key option instead.",
+      ),
+    );
+    process.exit(1);
+  }
+
+  const authJsonPath = path.join(loginHome, 'auth.json');
+  if (!fs.existsSync(authJsonPath)) {
+    removeLoginHome();
+    setupLog.step('auth', 'failed', durationMs, { PROVIDER: 'codex', METHOD: method, ERROR: 'auth_json_not_found' });
+    p.log.error(
+      brandBody('Codex login succeeded but no auth.json was written. Try again, or paste an API key instead.'),
+    );
+    process.exit(1);
+  }
+
+  try {
+    execFileSync(
+      'onecli',
+      [
+        'secrets',
+        'create',
+        '--name',
+        'Codex',
+        '--type',
+        'openai',
+        '--file',
+        authJsonPath,
+        '--host-pattern',
+        'chatgpt.com',
+      ],
+      { stdio: ['ignore', 'pipe', 'pipe'] },
+    );
+  } catch (err) {
+    removeLoginHome();
+    setupLog.step('auth', 'failed', durationMs, { PROVIDER: 'codex', METHOD: method, ERROR: String(err) });
+    p.log.error(
+      brandBody(
+        "Couldn't save your Codex credentials to the vault. Make sure OneCLI is running (`onecli version`), then retry.",
+      ),
+    );
+    process.exit(1);
+  }
+  removeLoginHome();
+  setupLog.step('auth', 'success', durationMs, { PROVIDER: 'codex', METHOD: method });
+  p.log.success(brandBody('OpenAI account connected — credentials live in your OneCLI vault, never in the container.'));
+}
+
+function runInherit(cmd: string, args: string[], extraEnv?: Record<string, string>): Promise<number> {
+  return new Promise((resolve) => {
+    const child = spawn(cmd, args, {
+      stdio: 'inherit',
+      env: extraEnv ? { ...process.env, ...extraEnv } : process.env,
+    });
+    child.on('close', (code) => resolve(code ?? 1));
+    child.on('error', () => resolve(1));
+  });
+}
+
+// ─── failure assist ──────────────────────────────────────────────────────
+
+/**
+ * The Codex CLI can debug a setup failure only if the binary runs AND
+ * ~/.codex/auth.json exists (API-key-only installs keep the key in the
+ * OneCLI vault, so the host-side CLI has nothing to authenticate with).
+ */
+export function isCodexCliUsable(): boolean {
+  const codexCheck = spawnSync('codex', ['--version'], { encoding: 'utf-8', stdio: ['ignore', 'pipe', 'pipe'] });
+  if (codexCheck.status !== 0) return false;
+  return fs.existsSync(path.join(os.homedir(), '.codex', 'auth.json'));
+}
+
+/**
+ * Failure prompt handed to the interactive Codex session — same content as
+ * the dispatcher's Claude system prompt: what failed, the job ("diagnose and
+ * fix, be concise, exit when done"), and a de-duped file reference list.
+ */
+export function buildCodexFailurePrompt(ctx: AssistContext, projectRoot: string): string {
+  const stepRefs = STEP_FILES[ctx.stepName] ?? [];
+  const references = [
+    ...BIG_PICTURE_FILES,
+    ...stepRefs,
+    'logs/setup.log',
+    ctx.rawLogPath ? path.relative(projectRoot, ctx.rawLogPath) : 'logs/setup-steps/',
+  ].filter((v, i, a) => a.indexOf(v) === i);
+
+  const lines: string[] = [
+    "The user is running NanoClaw's interactive setup flow and hit a failure.",
+    '',
+    `Failed step: ${ctx.stepName}`,
+    `Error: ${ctx.msg}`,
+  ];
+
+  if (ctx.hint) lines.push(`Hint: ${ctx.hint}`);
+
+  lines.push(
+    '',
+    'Your job: help them diagnose and fix this issue. Read the referenced files',
+    'and logs to understand what went wrong, then help them fix it. You can read',
+    'files, run commands, check logs, and explain what happened. Be concise.',
+    "When they're ready to resume setup, tell them to exit Codex.",
+    '',
+    'Relevant files (read as needed):',
+  );
+  for (const f of references) lines.push(`  - ${f}`);
+
+  return lines.join('\n');
+}
+
+/**
+ * Registry hook: offer to debug a setup failure with the Codex CLI. Returns
+ * 'unavailable' when the CLI can't run here so the dispatcher can fall back
+ * to its guarded Claude offer.
+ */
+export async function offerCodexFailureAssist(ctx: AssistContext, projectRoot: string): Promise<FailureAssistResult> {
+  if (!isCodexCliUsable()) return 'unavailable';
+
+  const want = ensureAnswer(
+    await p.confirm({
+      message: 'Want to debug this with Codex?',
+      initialValue: true,
+    }),
+  );
+  if (!want) return 'declined';
+
+  const prompt = buildCodexFailurePrompt(ctx, projectRoot);
+
+  note(
+    [
+      'Launching Codex to help debug this failure.',
+      'It has the context of what went wrong.',
+      '',
+      k.dim("Exit Codex (Ctrl-C or /quit) when you're ready to come back to setup."),
+    ].join('\n'),
+    'Handing off to Codex',
+  );
+
+  return new Promise<FailureAssistResult>((resolve) => {
+    // codex accepts a positional initial prompt for the interactive TUI.
+    const child = spawn('codex', [prompt], { cwd: projectRoot, stdio: 'inherit' });
+    child.on('close', () => {
+      p.log.success(brandBody("Back from Codex. Let's continue."));
+      resolve('launched');
+    });
+    child.on('error', () => {
+      p.log.error("Couldn't launch Codex.");
+      resolve('unavailable');
+    });
+  });
+}
+
+// ─── install verification ────────────────────────────────────────────────
+
+/**
+ * Verify the codex provider payload is fully wired — the same pre-flight the
+ * /add-codex skill checks. While codex ships in trunk these always pass; once
+ * the payload moves to the providers branch, a failed check means the install
+ * step should run (or the user finishes via /add-codex).
+ */
+export function verifyCodexInstall(): { ok: boolean; problems: string[] } {
+  const problems: string[] = [];
+  const root = process.cwd();
+
+  const requiredFiles = [
+    'src/providers/codex.ts',
+    'src/providers/codex-agents-md.ts',
+    'container/agent-runner/src/providers/codex.ts',
+    'container/agent-runner/src/providers/codex-app-server.ts',
+  ];
+  for (const file of requiredFiles) {
+    if (!fs.existsSync(path.join(root, file))) problems.push(`missing file: ${file}`);
+  }
+
+  for (const barrel of ['src/providers/index.ts', 'container/agent-runner/src/providers/index.ts']) {
+    const barrelPath = path.join(root, barrel);
+    if (!fs.existsSync(barrelPath) || !fs.readFileSync(barrelPath, 'utf-8').includes("import './codex.js';")) {
+      problems.push(`missing barrel import in ${barrel}`);
+    }
+  }
+
+  const manifestPath = path.join(root, 'container', 'cli-tools.json');
+  let hasCodexCli = false;
+  if (fs.existsSync(manifestPath)) {
+    try {
+      const tools = JSON.parse(fs.readFileSync(manifestPath, 'utf-8')) as Array<{ name?: string }>;
+      hasCodexCli = Array.isArray(tools) && tools.some((t) => t.name === '@openai/codex');
+    } catch {
+      hasCodexCli = false;
+    }
+  }
+  if (!hasCodexCli) {
+    problems.push('container/cli-tools.json missing the @openai/codex CLI entry');
+  }
+
+  return { ok: problems.length === 0, problems };
+}
+
+export async function runCodexInstallCheck(): Promise<void> {
+  p.log.step(brandBody('Checking the Codex provider install…'));
+  const { ok, problems } = verifyCodexInstall();
+  if (ok) {
+    setupLog.step('codex-install', 'success', 0, {});
+    p.log.success(brandBody('Codex installed properly.'));
+    return;
+  }
+
+  setupLog.step('codex-install', 'failed', 0, { PROBLEMS: problems.join('; ') });
+  p.log.warn(brandBody('The Codex provider is not fully installed:'));
+  for (const problem of problems) console.log(k.dim(`   • ${problem}`));
+  p.log.warn(
+    brandBody(
+      'Finish it with your coding agent of choice: open Codex CLI or Claude Code in this repo and run the /add-codex skill. Setup will continue — Codex groups will work once the install completes.',
+    ),
+  );
+}
+
+// Self-registration: the setup picker and the standalone `provider-auth` step
+// render from the registry — this call is codex's only reach-in to the setup
+// flow (guarded by the barrel-driven registration test).
+registerSetupProvider({
+  value: 'codex',
+  label: 'Codex',
+  hint: 'OpenAI — ChatGPT subscription or API key',
+  runAuth: runCodexAuthStep,
+  runInstallCheck: runCodexInstallCheck,
+  offerFailureAssist: offerCodexFailureAssist,
+});

setup/register.ts

@@ -20,6 +20,7 @@ import {
 import { isValidGroupFolder } from '../src/group-folder.js';
 import { initGroupFilesystem } from '../src/group-init.js';
 import { log } from '../src/log.js';
+import { namespacedPlatformId } from '../src/platform-id.js';
 import { resolveSession, writeSessionMessage } from '../src/session-manager.js';
 import { emitStatus } from './status.js';
 
@@ -112,12 +113,10 @@ export async function run(args: string[]): Promise<void> {
     process.exit(4);
   }
 
-  // Chat SDK adapters prefix platform IDs with the channel type
-  // (e.g. "telegram:123", "discord:guild:channel"). Normalize here so
-  // the stored ID always matches what the adapter sends at runtime.
-  if (!parsed.platformId.startsWith(`${parsed.channel}:`)) {
-    parsed.platformId = `${parsed.channel}:${parsed.platformId}`;
-  }
+  // Normalize platform_id to the same shape the adapter will emit at runtime,
+  // so the router's (channel_type, platform_id) lookup matches what we store.
+  // Chat SDK adapters prefix, native adapters (WhatsApp/iMessage/Signal) don't.
+  parsed.platformId = namespacedPlatformId(parsed.channel, parsed.platformId);
 
   log.info('Registering channel', parsed);
 
@@ -167,19 +166,22 @@ export async function run(args: string[]): Promise<void> {
   if (!existing) {
     newlyWired = true;
     const mgaId = generateId('mga');
-    const triggerRules = parsed.trigger
-      ? JSON.stringify({
-          pattern: parsed.trigger,
-          requiresTrigger: parsed.requiresTrigger,
-        })
-      : null;
+    // Mirrors scripts/init-first-agent.ts:wireIfMissing so both setup paths
+    // create rows with the same shape. Groups default to 'mention' (bot only
+    // responds when addressed); DMs default to 'pattern'/'.' (respond to
+    // every message). An explicit --trigger overrides the pattern regex.
+    const isGroup = messagingGroup.is_group === 1;
+    const engageMode: 'pattern' | 'mention' = isGroup && !parsed.trigger ? 'mention' : 'pattern';
+    const engagePattern: string | null = engageMode === 'pattern' ? parsed.trigger || '.' : null;
     createMessagingGroupAgent({
       id: mgaId,
       messaging_group_id: messagingGroup.id,
       agent_group_id: agentGroup.id,
-      trigger_rules: triggerRules,
-      response_scope: 'all',
-      session_mode: parsed.sessionMode,
+      engage_mode: engageMode,
+      engage_pattern: engagePattern,
+      sender_scope: 'all',
+      ignored_message_policy: 'drop',
+      session_mode: parsed.sessionMode as 'shared' | 'per-thread' | 'agent-shared',
       priority: 0,
       created_at: new Date().toISOString(),
     });

setup/service.ts

@@ -11,6 +11,7 @@ import path from 'path';
 
 import { log } from '../src/log.js';
 import { getLaunchdLabel, getSystemdUnit } from '../src/install-slug.js';
+import { cleanupUnhealthyPeers } from './peer-cleanup.js';
 import {
   commandExists,
   getPlatform,
@@ -53,6 +54,19 @@ export async function run(_args: string[]): Promise<void> {
 
   fs.mkdirSync(path.join(projectRoot, 'logs'), { recursive: true });
 
+  // Peer preflight — a crash-looping peer install (most often the legacy v1
+  // `com.nanoclaw` plist) will keep trashing this install's containers on
+  // every respawn via its own cleanupOrphans. Detect and unload any peer
+  // that's unhealthy before we install our service. Healthy peers are left
+  // alone now that container reaping is install-label-scoped.
+  const peerReport = cleanupUnhealthyPeers(projectRoot);
+  if (peerReport.unloaded.length > 0) {
+    log.warn('Unloaded unhealthy peer NanoClaw services', {
+      count: peerReport.unloaded.length,
+      labels: peerReport.unloaded.map((p) => p.label),
+    });
+  }
+
   if (platform === 'macos') {
     setupLaunchd(projectRoot, nodePath, homeDir);
   } else if (platform === 'linux') {

setup/signal-auth.ts

@@ -0,0 +1,182 @@
+/**
+ * Step: signal-auth — link this host to an existing Signal account via
+ * signal-cli's QR-code flow.
+ *
+ * signal-cli `link` opens a bi-directional handshake with the Signal
+ * servers: it prints one line containing a linking URL (`sgnl://linkdevice?…`
+ * or older `tsdevice://linkdevice?…`), then blocks until either the user
+ * scans it from an existing Signal install, or the code expires. On
+ * success, a secondary account is created under the user's signal-cli
+ * data directory, associated with the phone number of the scanner.
+ *
+ * Methods:
+ *   (no args)                    Spawn signal-cli link, emit SIGNAL_AUTH_QR
+ *                                with the URL, wait for completion.
+ *
+ * Block schema (parent parses these):
+ *   SIGNAL_AUTH_QR       { QR: "<sgnl:// or tsdevice:// url>" }   — one-shot
+ *   SIGNAL_AUTH          { STATUS: success, ACCOUNT: +<digits> }  — terminal
+ *                        { STATUS: skipped, ACCOUNT, REASON: already-authenticated }
+ *                        { STATUS: failed, ERROR: <reason> }
+ *
+ * STATUS values match the runner's vocabulary (success/skipped/failed) so
+ * spawnStep recognises them and sets `ok` correctly; Signal-specific UI
+ * lives in setup/channels/signal.ts.
+ *
+ * If one or more accounts are already linked (discovered via
+ * `signal-cli -o json listAccounts`), the step emits SIGNAL_AUTH
+ * STATUS=skipped with the first account so the driver can reuse it.
+ * Selecting a different existing account is a driver concern.
+ */
+import { spawn, spawnSync } from 'child_process';
+
+import { emitStatus } from './status.js';
+
+const LINK_TIMEOUT_MS = 180_000;
+const DEFAULT_DEVICE_NAME = 'NanoClaw';
+
+interface SignalAccount {
+  account?: string;
+  registered?: boolean;
+}
+
+function cliPath(): string {
+  return process.env.SIGNAL_CLI_PATH || 'signal-cli';
+}
+
+/**
+ * Query signal-cli for currently linked accounts. Empty array if none
+ * configured, no binary, or the call fails for any other reason.
+ */
+function listAccounts(): string[] {
+  const cli = cliPath();
+  try {
+    const res = spawnSync(cli, ['-o', 'json', 'listAccounts'], {
+      encoding: 'utf-8',
+      stdio: ['ignore', 'pipe', 'pipe'],
+    });
+    if (res.status !== 0) return [];
+    const parsed = JSON.parse(res.stdout || '[]') as SignalAccount[];
+    return parsed
+      .filter((a) => a.registered !== false)
+      .map((a) => a.account ?? '')
+      .filter(Boolean);
+  } catch {
+    return [];
+  }
+}
+
+export async function run(_args: string[]): Promise<void> {
+  const cli = cliPath();
+
+  // Verify signal-cli exists before we commit to the long-running link.
+  // The driver checks too, but this keeps the step honest when run alone.
+  const probe = spawnSync(cli, ['--version'], {
+    stdio: ['ignore', 'pipe', 'pipe'],
+  });
+  if (probe.error || probe.status !== 0) {
+    emitStatus('SIGNAL_AUTH', {
+      STATUS: 'failed',
+      ERROR: 'signal-cli not found. Install signal-cli first.',
+    });
+    return;
+  }
+
+  const existing = listAccounts();
+  if (existing.length > 0) {
+    emitStatus('SIGNAL_AUTH', {
+      STATUS: 'skipped',
+      ACCOUNT: existing[0],
+      REASON: 'already-authenticated',
+    });
+    return;
+  }
+
+  await new Promise<void>((resolve) => {
+    let settled = false;
+    let qrEmitted = false;
+
+    const finish = (block: Record<string, string | number | boolean>, code: number): void => {
+      if (settled) return;
+      settled = true;
+      clearTimeout(timer);
+      emitStatus('SIGNAL_AUTH', block);
+      resolve();
+      setTimeout(() => process.exit(code), 500);
+    };
+
+    const timer = setTimeout(() => {
+      try {
+        child.kill('SIGTERM');
+      } catch {
+        /* ignore */
+      }
+      finish({ STATUS: 'failed', ERROR: 'qr_timeout' }, 1);
+    }, LINK_TIMEOUT_MS);
+
+    const child = spawn(cli, ['link', '--name', DEFAULT_DEVICE_NAME], {
+      stdio: ['ignore', 'pipe', 'pipe'],
+    });
+
+    // stdout carries the URL on the first line; subsequent lines may print
+    // status like "Associated with: +1555…". We don't strictly need to parse
+    // the number — listAccounts after exit is the source of truth — but the
+    // URL match drives the QR emit, which is the whole point.
+    let stdoutBuf = '';
+    const handleStdout = (chunk: Buffer): void => {
+      stdoutBuf += chunk.toString('utf-8');
+      let idx: number;
+      while ((idx = stdoutBuf.indexOf('\n')) !== -1) {
+        const line = stdoutBuf.slice(0, idx).trim();
+        stdoutBuf = stdoutBuf.slice(idx + 1);
+        if (!line) continue;
+        // Match both modern (sgnl://) and legacy (tsdevice://) schemes.
+        if (/^(sgnl|tsdevice):\/\/linkdevice\?/.test(line) && !qrEmitted) {
+          qrEmitted = true;
+          emitStatus('SIGNAL_AUTH_QR', { QR: line });
+        }
+      }
+    };
+    child.stdout.on('data', handleStdout);
+
+    // Capture stderr for the transcript / log — signal-cli writes warnings
+    // and errors there. We don't emit on partial stderr lines since a
+    // successful link can still produce noise.
+    let stderrBuf = '';
+    child.stderr.on('data', (chunk: Buffer) => {
+      stderrBuf += chunk.toString('utf-8');
+    });
+
+    child.on('error', (err) => {
+      finish({ STATUS: 'failed', ERROR: `spawn error: ${err.message}` }, 1);
+    });
+
+    child.on('close', (code) => {
+      // After a successful link, signal-cli exits 0 and the newly linked
+      // account shows up in listAccounts. Use that as the source of truth
+      // rather than scraping stdout — more robust across signal-cli versions.
+      if (code === 0) {
+        const post = listAccounts();
+        if (post.length === 0) {
+          finish(
+            { STATUS: 'failed', ERROR: 'link exited 0 but no account registered' },
+            1,
+          );
+          return;
+        }
+        finish({ STATUS: 'success', ACCOUNT: post[0] }, 0);
+        return;
+      }
+
+      // Non-zero exit. Surface the last non-empty stderr line for context;
+      // signal-cli's own error messages are usually informative.
+      const lastErr =
+        stderrBuf
+          .split('\n')
+          .map((l) => l.trim())
+          .filter(Boolean)
+          .slice(-1)[0] ?? `signal-cli link exited with code ${code}`;
+      finish({ STATUS: 'failed', ERROR: lastErr }, 1);
+    });
+  });
+}

setup/verify.test.ts

@@ -0,0 +1,55 @@
+import { describe, expect, it } from 'vitest';
+
+import { determineVerifyStatus } from './verify.js';
+
+const healthyBase = {
+  service: 'running' as const,
+  credentials: 'configured',
+  anyChannelConfigured: false,
+  registeredGroups: 1,
+  agentPing: 'ok' as const,
+};
+
+describe('determineVerifyStatus', () => {
+  it('accepts a working CLI-only install', () => {
+    expect(determineVerifyStatus(healthyBase)).toBe('success');
+  });
+
+  it('accepts a messaging-channel install when CLI ping is skipped', () => {
+    expect(
+      determineVerifyStatus({
+        ...healthyBase,
+        anyChannelConfigured: true,
+        agentPing: 'skipped',
+      }),
+    ).toBe('success');
+  });
+
+  it('fails when neither CLI nor messaging channels are usable', () => {
+    expect(
+      determineVerifyStatus({
+        ...healthyBase,
+        agentPing: 'skipped',
+      }),
+    ).toBe('failed');
+  });
+
+  it('fails when the CLI agent does not respond', () => {
+    expect(
+      determineVerifyStatus({
+        ...healthyBase,
+        anyChannelConfigured: true,
+        agentPing: 'no_reply',
+      }),
+    ).toBe('failed');
+  });
+
+  it('fails when no agent groups are registered', () => {
+    expect(
+      determineVerifyStatus({
+        ...healthyBase,
+        registeredGroups: 0,
+      }),
+    ).toBe('failed');
+  });
+});

setup/verify.ts

@@ -14,7 +14,7 @@ import Database from 'better-sqlite3';
 import { DATA_DIR } from '../src/config.js';
 import { readEnvFile } from '../src/env.js';
 import { log } from '../src/log.js';
-import { pingCliAgent } from './lib/agent-ping.js';
+import { pingCliAgent, type PingResult } from './lib/agent-ping.js';
 import { getLaunchdLabel, getSystemdUnit } from '../src/install-slug.js';
 import {
   getPlatform,
@@ -220,22 +220,22 @@ export async function run(_args: string[]): Promise<void> {
 
   // 7. End-to-end: ping the CLI agent and confirm it replies. Only run if
   // everything upstream looks healthy, since a broken socket would just hang.
-  let agentPing: 'ok' | 'no_reply' | 'socket_error' | 'skipped' = 'skipped';
+  let agentPing: 'ok' | 'no_reply' | 'socket_error' | 'auth_error' | 'skipped' = 'skipped';
   if (service === 'running' && registeredGroups > 0) {
     log.info('Pinging CLI agent');
     agentPing = await pingCliAgent();
     log.info('Agent ping result', { agentPing });
   }
 
-  // Determine overall status
-  const status =
-    service === 'running' &&
-    credentials !== 'missing' &&
-    anyChannelConfigured &&
-    registeredGroups > 0 &&
-    (agentPing === 'ok' || agentPing === 'skipped')
-      ? 'success'
-      : 'failed';
+  // Determine overall status. A CLI-only install is valid when the local
+  // agent round-trip succeeds; messaging app credentials are optional.
+  const status = determineVerifyStatus({
+    service,
+    credentials,
+    anyChannelConfigured,
+    registeredGroups,
+    agentPing,
+  });
 
   log.info('Verification complete', { status, channelAuth });
 
@@ -255,6 +255,25 @@ export async function run(_args: string[]): Promise<void> {
   if (status === 'failed') process.exit(1);
 }
 
+export function determineVerifyStatus(input: {
+  service: 'not_found' | 'stopped' | 'running' | 'running_other_checkout';
+  credentials: string;
+  anyChannelConfigured: boolean;
+  registeredGroups: number;
+  agentPing: PingResult | 'skipped';
+}): 'success' | 'failed' {
+  const cliAgentResponds = input.agentPing === 'ok';
+  const hasUsableChannel = input.anyChannelConfigured || cliAgentResponds;
+
+  return input.service === 'running' &&
+    input.credentials !== 'missing' &&
+    hasUsableChannel &&
+    input.registeredGroups > 0 &&
+    (cliAgentResponds || input.agentPing === 'skipped')
+    ? 'success'
+    : 'failed';
+}
+
 /**
  * Given a PID, resolve the script path the process is executing (i.e. the
  * first `.js` / `.ts` / `.mjs` arg after `node`). Returns null on any

src/attachment-safety.ts

@@ -0,0 +1,23 @@
+import path from 'path';
+
+/**
+ * Is `name` safe to use as the last segment of a path inside an
+ * attachment-staging directory? Filenames originate from untrusted sources —
+ * channel messages from any chat participant, agent-to-agent forwards from
+ * a possibly-compromised peer agent — and land in `path.join(dir, name)`
+ * sinks on the host. Without this guard, a `..`-laden name escapes the
+ * inbox and writes anywhere the host process has filesystem permission.
+ *
+ * Rejects:
+ *   - non-string / empty
+ *   - `.` / `..` (traversal sentinels that path.basename returns as-is)
+ *   - anything containing a path separator (`/` or `\`) or NUL
+ *   - any value where `path.basename(name) !== name`, catching OS-specific
+ *     separators and covering drives/prefixes on Windows runtimes
+ */
+export function isSafeAttachmentName(name: string): boolean {
+  if (typeof name !== 'string' || name.length === 0) return false;
+  if (name === '.' || name === '..') return false;
+  if (/[\\/\0]/.test(name)) return false;
+  return path.basename(name) === name;
+}

src/channels/adapter.ts

@@ -56,6 +56,8 @@ export interface InboundEvent {
      * See InboundMessage.isMention for the full explanation.
      */
     isMention?: boolean;
+    /** True when the source is a group/channel thread, false for DMs. */
+    isGroup?: boolean;
   };
   replyTo?: DeliveryAddress;
 }
@@ -81,6 +83,8 @@ export interface InboundMessage {
    * router falls back to text-match against agent_group_name.
    */
   isMention?: boolean;
+  /** True when the source is a group/channel thread, false for DMs. */
+  isGroup?: boolean;
 }
 
 /** A file attachment to deliver alongside a message. */

src/channels/chat-sdk-bridge.ts

@@ -81,6 +81,26 @@ export interface ChatSdkBridgeConfig {
  * chunk boundary will render as two independent blocks on the receiving
  * platform, which is the same behavior as manually re-opening a fence.
  */
+/**
+ * Decode the actual option value from a button callback. Buttons are encoded
+ * with an integer index (to keep under Telegram's 64-byte callback_data cap),
+ * and the real value is looked up via `getAskQuestionRender(questionId)`.
+ * Falls back to treating the tail as a literal value so old in-flight cards
+ * (encoded before this shortening landed) still resolve.
+ */
+function resolveSelectedOption(
+  render: { options: NormalizedOption[] } | undefined,
+  eventValue: string | undefined,
+  tail: string | undefined,
+): string {
+  const candidate = eventValue ?? tail ?? '';
+  if (render && /^\d+$/.test(candidate)) {
+    const idx = Number(candidate);
+    if (render.options[idx]) return render.options[idx].value;
+  }
+  return candidate;
+}
+
 export function splitForLimit(text: string, limit: number): string[] {
   if (text.length <= limit) return [text];
   const chunks: string[] = [];
@@ -105,7 +125,11 @@ export function createChatSdkBridge(config: ChatSdkBridgeConfig): ChannelAdapter
   let setupConfig: ChannelSetup;
   let gatewayAbort: AbortController | null = null;
 
-  async function messageToInbound(message: ChatMessage, isMention: boolean): Promise<InboundMessage> {
+  async function messageToInbound(
+    message: ChatMessage,
+    isMention: boolean,
+    isGroup?: boolean,
+  ): Promise<InboundMessage> {
     // eslint-disable-next-line @typescript-eslint/no-explicit-any
     const serialized = message.toJSON() as Record<string, any>;
 
@@ -162,6 +186,7 @@ export function createChatSdkBridge(config: ChatSdkBridgeConfig): ChannelAdapter
       content: serialized,
       timestamp: message.metadata.dateSent.toISOString(),
       isMention,
+      isGroup,
     };
   }
 
@@ -195,13 +220,17 @@ export function createChatSdkBridge(config: ChatSdkBridgeConfig): ChannelAdapter
       // wirings still fire on in-thread mentions.
       chat.onSubscribedMessage(async (thread, message) => {
         const channelId = adapter.channelIdFromThreadId(thread.id);
-        await setupConfig.onInbound(channelId, thread.id, await messageToInbound(message, message.isMention === true));
+        await setupConfig.onInbound(
+          channelId,
+          thread.id,
+          await messageToInbound(message, message.isMention === true, true),
+        );
       });
 
       // @mention in an unsubscribed thread — SDK-confirmed bot mention.
       chat.onNewMention(async (thread, message) => {
         const channelId = adapter.channelIdFromThreadId(thread.id);
-        await setupConfig.onInbound(channelId, thread.id, await messageToInbound(message, true));
+        await setupConfig.onInbound(channelId, thread.id, await messageToInbound(message, true, true));
       });
 
       // DMs — by definition addressed to the bot. Thread id flows through
@@ -216,7 +245,7 @@ export function createChatSdkBridge(config: ChatSdkBridgeConfig): ChannelAdapter
           sender: (message.author as any)?.fullName ?? (message.author as any)?.userId ?? 'unknown',
           threadId: thread.id,
         });
-        await setupConfig.onInbound(channelId, thread.id, await messageToInbound(message, true));
+        await setupConfig.onInbound(channelId, thread.id, await messageToInbound(message, true, false));
       });
 
       // Plain messages in unsubscribed threads.
@@ -231,7 +260,7 @@ export function createChatSdkBridge(config: ChatSdkBridgeConfig): ChannelAdapter
       // flood gate.
       chat.onNewMessage(/./, async (thread, message) => {
         const channelId = adapter.channelIdFromThreadId(thread.id);
-        await setupConfig.onInbound(channelId, thread.id, await messageToInbound(message, false));
+        await setupConfig.onInbound(channelId, thread.id, await messageToInbound(message, false, true));
       });
 
       // Handle button clicks (ask_user_question)
@@ -240,11 +269,15 @@ export function createChatSdkBridge(config: ChatSdkBridgeConfig): ChannelAdapter
         const parts = event.actionId.split(':');
         if (parts.length < 3) return;
         const questionId = parts[1];
-        const selectedOption = event.value || '';
+        const tail = parts.slice(2).join(':');
         const userId = event.user?.userId || '';
 
         // Resolve render metadata BEFORE dispatching onAction (which deletes the row).
         const render = getAskQuestionRender(questionId);
+        // New format: button id/value is an integer index into options (kept
+        // short to fit Telegram's 64-byte callback_data cap). Old format:
+        // the full value is embedded in actionId/value directly.
+        const selectedOption = resolveSelectedOption(render, event.value, tail);
         const title = render?.title ?? '❓ Question';
         const matched = render?.options.find((o) => o.value === selectedOption);
         const selectedLabel = matched?.selectedLabel ?? selectedOption ?? '(clicked)';
@@ -348,8 +381,13 @@ export function createChatSdkBridge(config: ChatSdkBridgeConfig): ChannelAdapter
           children: [
             CardText(question),
             Actions(
-              options.map((opt) =>
-                Button({ id: `ncq:${questionId}:${opt.value}`, label: opt.label, value: opt.value }),
+              // Encode button id/value with the option index rather than the
+              // full value. Telegram caps callback_data at 64 bytes, and
+              // long values (e.g. ISO datetimes, URLs) push the JSON payload
+              // well past that. The onAction handlers resolve the index back
+              // to the real value via getAskQuestionRender(questionId).
+              options.map((opt, idx) =>
+                Button({ id: `ncq:${questionId}:${idx}`, label: opt.label, value: String(idx) }),
               ),
             ),
           ],
@@ -501,18 +539,21 @@ async function handleForwardedEvent(
     // type 3 = MessageComponent (button/select)
     if (interaction.type === 3) {
       const customId = (interaction.data as Record<string, unknown>)?.custom_id as string;
-      const user = (interaction.member as Record<string, unknown>)?.user as Record<string, string> | undefined;
+      // In guilds the clicker is at interaction.member.user; in DMs it's interaction.user directly.
+      const user =
+        ((interaction.member as Record<string, unknown>)?.user as Record<string, string> | undefined) ??
+        (interaction.user as Record<string, string> | undefined);
       const interactionId = interaction.id as string;
       const interactionToken = interaction.token as string;
 
       // Parse the selected option from custom_id
       let questionId: string | undefined;
-      let selectedOption: string | undefined;
+      let tail: string | undefined;
       if (customId?.startsWith('ncq:')) {
         const colonIdx = customId.indexOf(':', 4); // after "ncq:"
         if (colonIdx !== -1) {
           questionId = customId.slice(4, colonIdx);
-          selectedOption = customId.slice(colonIdx + 1);
+          tail = customId.slice(colonIdx + 1);
         }
       }
 
@@ -521,6 +562,9 @@ async function handleForwardedEvent(
         ((interaction.message as Record<string, unknown>)?.embeds as Array<Record<string, unknown>>) || [];
       const originalDescription = (originalEmbeds[0]?.description as string) || '';
       const render = questionId ? getAskQuestionRender(questionId) : undefined;
+      // Discord custom_id mirrors the new index-based encoding (see Button
+      // construction). Decode back to the real option value for downstream.
+      const selectedOption = resolveSelectedOption(render, tail, tail);
       const cardTitle = render?.title ?? ((originalEmbeds[0]?.title as string) || '❓ Question');
       const matchedOpt = render?.options.find((o) => o.value === selectedOption);
       const selectedLabel = matchedOpt?.selectedLabel ?? selectedOption ?? customId;

src/config.ts

@@ -2,7 +2,7 @@ import os from 'os';
 import path from 'path';
 
 import { readEnvFile } from './env.js';
-import { getContainerImageBase, getDefaultContainerImage } from './install-slug.js';
+import { getContainerImageBase, getDefaultContainerImage, getInstallSlug } from './install-slug.js';
 import { isValidTimezone } from './timezone.js';
 
 // Read config values from .env (falls back to process.env).
@@ -27,6 +27,10 @@ export const DATA_DIR = path.resolve(PROJECT_ROOT, 'data');
 // `nanoclaw-agent:latest` and clobber each other on rebuild.
 export const CONTAINER_IMAGE_BASE = process.env.CONTAINER_IMAGE_BASE || getContainerImageBase(PROJECT_ROOT);
 export const CONTAINER_IMAGE = process.env.CONTAINER_IMAGE || getDefaultContainerImage(PROJECT_ROOT);
+// Install slug — stamped onto every spawned container via --label so
+// cleanupOrphans only reaps containers from this install, not peers.
+export const INSTALL_SLUG = getInstallSlug(PROJECT_ROOT);
+export const CONTAINER_INSTALL_LABEL = `nanoclaw-install=${INSTALL_SLUG}`;
 export const CONTAINER_TIMEOUT = parseInt(process.env.CONTAINER_TIMEOUT || '1800000', 10);
 export const CONTAINER_MAX_OUTPUT_SIZE = parseInt(process.env.CONTAINER_MAX_OUTPUT_SIZE || '10485760', 10); // 10MB default
 export const ONECLI_URL = process.env.ONECLI_URL || envConfig.ONECLI_URL;

src/container-runner.test.ts

@@ -0,0 +1,32 @@
+import { describe, expect, it } from 'vitest';
+
+import { resolveProviderName } from './container-runner.js';
+
+describe('resolveProviderName', () => {
+  it('prefers session over group and container.json', () => {
+    expect(resolveProviderName('codex', 'opencode', 'claude')).toBe('codex');
+  });
+
+  it('falls back to group when session is null', () => {
+    expect(resolveProviderName(null, 'codex', 'claude')).toBe('codex');
+  });
+
+  it('falls back to container.json when session and group are null', () => {
+    expect(resolveProviderName(null, null, 'opencode')).toBe('opencode');
+  });
+
+  it('defaults to claude when nothing is set', () => {
+    expect(resolveProviderName(null, null, undefined)).toBe('claude');
+  });
+
+  it('lowercases the resolved name', () => {
+    expect(resolveProviderName('CODEX', null, null)).toBe('codex');
+    expect(resolveProviderName(null, 'OpenCode', null)).toBe('opencode');
+    expect(resolveProviderName(null, null, 'Claude')).toBe('claude');
+  });
+
+  it('treats empty string as unset (falls through)', () => {
+    expect(resolveProviderName('', 'codex', null)).toBe('codex');
+    expect(resolveProviderName(null, '', 'opencode')).toBe('opencode');
+  });
+});

src/container-runner.ts

@@ -12,6 +12,7 @@ import { OneCLI } from '@onecli-sh/sdk';
 import {
   CONTAINER_IMAGE,
   CONTAINER_IMAGE_BASE,
+  CONTAINER_INSTALL_LABEL,
   DATA_DIR,
   GROUPS_DIR,
   ONECLI_API_KEY,
@@ -35,7 +36,13 @@ import {
   type ProviderContainerContribution,
   type VolumeMount,
 } from './providers/provider-container-registry.js';
-import { markContainerRunning, markContainerStopped, sessionDir, writeSessionRouting } from './session-manager.js';
+import {
+  heartbeatPath,
+  markContainerRunning,
+  markContainerStopped,
+  sessionDir,
+  writeSessionRouting,
+} from './session-manager.js';
 import type { AgentGroup, Session } from './types.js';
 
 const onecli = new OneCLI({ url: ONECLI_URL, apiKey: ONECLI_API_KEY });
@@ -130,6 +137,12 @@ async function spawnContainer(session: Session): Promise<void> {
 
   log.info('Spawning container', { sessionId: session.id, agentGroup: agentGroup.name, containerName });
 
+  // Clear any orphan heartbeat from a previous container instance — the
+  // sweep's ceiling check treats a missing file as "fresh spawn, give grace"
+  // (host-sweep.ts line 87). Without this, the stale mtime can trigger an
+  // immediate kill before the new container touches the file itself.
+  fs.rmSync(heartbeatPath(agentGroup.id, session.id), { force: true });
+
   const container = spawn(CONTAINER_RUNTIME_BIN, args, { stdio: ['ignore', 'pipe', 'pipe'] });
 
   activeContainers.set(session.id, { process: container, containerName });
@@ -178,12 +191,31 @@ export function killContainer(sessionId: string, reason: string): void {
   }
 }
 
+/**
+ * Resolve the provider name for a session using the precedence documented in
+ * the provider-install skills:
+ *
+ *   sessions.agent_provider
+ *     → agent_groups.agent_provider
+ *     → container.json `provider`
+ *     → 'claude'
+ *
+ * Pure so the precedence can be unit-tested without a DB or filesystem.
+ */
+export function resolveProviderName(
+  sessionProvider: string | null | undefined,
+  agentGroupProvider: string | null | undefined,
+  containerConfigProvider: string | null | undefined,
+): string {
+  return (sessionProvider || agentGroupProvider || containerConfigProvider || 'claude').toLowerCase();
+}
+
 function resolveProviderContribution(
   session: Session,
   agentGroup: AgentGroup,
   containerConfig: import('./container-config.js').ContainerConfig,
 ): { provider: string; contribution: ProviderContainerContribution } {
-  const provider = (containerConfig.provider || 'claude').toLowerCase();
+  const provider = resolveProviderName(session.agent_provider, agentGroup.agent_provider, containerConfig.provider);
   const fn = getProviderContainerConfig(provider);
   const contribution = fn
     ? fn({
@@ -389,7 +421,7 @@ async function buildContainerArgs(
   providerContribution: ProviderContainerContribution,
   agentIdentifier?: string,
 ): Promise<string[]> {
-  const args: string[] = ['run', '--rm', '--name', containerName];
+  const args: string[] = ['run', '--rm', '--name', containerName, '--label', CONTAINER_INSTALL_LABEL];
 
   // Environment — only vars read by code we don't own.
   // Everything NanoClaw-specific is in container.json (read by runner at startup).

src/container-runtime.test.ts

@@ -24,6 +24,7 @@ import {
   ensureContainerRuntimeRunning,
   cleanupOrphans,
 } from './container-runtime.js';
+import { CONTAINER_INSTALL_LABEL } from './config.js';
 import { log } from './log.js';
 
 beforeEach(() => {
@@ -84,6 +85,17 @@ describe('ensureContainerRuntimeRunning', () => {
 // --- cleanupOrphans ---
 
 describe('cleanupOrphans', () => {
+  it('filters ps by the install label so peers are not reaped', () => {
+    mockExecSync.mockReturnValueOnce('');
+
+    cleanupOrphans();
+
+    expect(mockExecSync).toHaveBeenCalledWith(
+      `${CONTAINER_RUNTIME_BIN} ps --filter label=${CONTAINER_INSTALL_LABEL} --format '{{.Names}}'`,
+      expect.any(Object),
+    );
+  });
+
   it('stops orphaned nanoclaw containers', () => {
     // docker ps returns container names, one per line
     mockExecSync.mockReturnValueOnce('nanoclaw-group1-111\nnanoclaw-group2-222\n');

src/container-runtime.ts

@@ -5,6 +5,7 @@
 import { execSync } from 'child_process';
 import os from 'os';
 
+import { CONTAINER_INSTALL_LABEL } from './config.js';
 import { log } from './log.js';
 
 /** The container runtime binary name. */
@@ -56,13 +57,22 @@ export function ensureContainerRuntimeRunning(): void {
   }
 }
 
-/** Kill orphaned NanoClaw containers from previous runs. */
+/**
+ * Kill orphaned NanoClaw containers from THIS install's previous runs.
+ *
+ * Scoped by label `nanoclaw-install=<slug>` so a crash-looping peer install
+ * cannot reap our containers, and we cannot reap theirs. The label is
+ * stamped onto every container at spawn time — see container-runner.ts.
+ */
 export function cleanupOrphans(): void {
   try {
-    const output = execSync(`${CONTAINER_RUNTIME_BIN} ps --filter name=nanoclaw- --format '{{.Names}}'`, {
-      stdio: ['pipe', 'pipe', 'pipe'],
-      encoding: 'utf-8',
-    });
+    const output = execSync(
+      `${CONTAINER_RUNTIME_BIN} ps --filter label=${CONTAINER_INSTALL_LABEL} --format '{{.Names}}'`,
+      {
+        stdio: ['pipe', 'pipe', 'pipe'],
+        encoding: 'utf-8',
+      },
+    );
     const orphans = output.trim().split('\n').filter(Boolean);
     for (const name of orphans) {
       try {

src/db/migrations/013-approval-render-metadata.ts

@@ -0,0 +1,27 @@
+/**
+ * Persist ask_question render metadata (title + options_json) on
+ * `pending_channel_approvals` and `pending_sender_approvals`, mirroring the
+ * columns migration 003 / module-approvals-title-options added to
+ * `pending_approvals`.
+ *
+ * Before this, `getAskQuestionRender` hardcoded the title + option labels
+ * for these two tables in the DB-access layer — duplicating wording that
+ * also lived in the approval modules and causing a visible drift between
+ * the initial card title ("📣 Bot mentioned in new chat" / "💬 New direct
+ * message", chosen per event) and the post-click render ("📣 Channel
+ * registration", constant). Storing the render metadata alongside the row
+ * lets both sides read from the same source.
+ */
+import type Database from 'better-sqlite3';
+import type { Migration } from './index.js';
+
+export const migration013: Migration = {
+  version: 13,
+  name: 'approval-render-metadata',
+  up(db: Database.Database) {
+    db.exec(`ALTER TABLE pending_channel_approvals ADD COLUMN title TEXT NOT NULL DEFAULT ''`);
+    db.exec(`ALTER TABLE pending_channel_approvals ADD COLUMN options_json TEXT NOT NULL DEFAULT '[]'`);
+    db.exec(`ALTER TABLE pending_sender_approvals ADD COLUMN title TEXT NOT NULL DEFAULT ''`);
+    db.exec(`ALTER TABLE pending_sender_approvals ADD COLUMN options_json TEXT NOT NULL DEFAULT '[]'`);
+  },
+};

src/db/migrations/index.ts

@@ -9,6 +9,7 @@ import { migration009 } from './009-drop-pending-credentials.js';
 import { migration010 } from './010-engage-modes.js';
 import { migration011 } from './011-pending-sender-approvals.js';
 import { migration012 } from './012-channel-registration.js';
+import { migration013 } from './013-approval-render-metadata.js';
 import { moduleApprovalsPendingApprovals } from './module-approvals-pending-approvals.js';
 import { moduleApprovalsTitleOptions } from './module-approvals-title-options.js';
 
@@ -29,6 +30,7 @@ const migrations: Migration[] = [
   migration010,
   migration011,
   migration012,
+  migration013,
 ];
 
 export function runMigrations(db: Database.Database): void {

src/db/session-db.ts

@@ -139,10 +139,10 @@ export function getMessageForRetry(
   db: Database.Database,
   messageId: string,
   status: string,
-): { id: string; tries: number } | undefined {
-  return db.prepare('SELECT id, tries FROM messages_in WHERE id = ? AND status = ?').get(messageId, status) as
-    | { id: string; tries: number }
-    | undefined;
+): { id: string; tries: number; processAfter: string | null } | undefined {
+  return db
+    .prepare('SELECT id, tries, process_after as processAfter FROM messages_in WHERE id = ? AND status = ?')
+    .get(messageId, status) as { id: string; tries: number; processAfter: string | null } | undefined;
 }
 
 export function syncProcessingAcks(inDb: Database.Database, outDb: Database.Database): void {

src/db/sessions.ts

@@ -1,5 +1,5 @@
 import type { PendingApproval, PendingQuestion, Session } from '../types.js';
-import { getDb } from './connection.js';
+import { getDb, hasTable } from './connection.js';
 
 // ── Sessions ──
 
@@ -97,10 +97,16 @@ export function deleteSession(id: string): void {
 
 // ── Pending Questions ──
 
-export function createPendingQuestion(pq: PendingQuestion): void {
-  getDb()
+/**
+ * Insert a pending question row. Idempotent: when delivery fails and retries,
+ * the second attempt calls this with the same question_id — without `OR
+ * IGNORE` that would throw UNIQUE and prevent the retry from reaching the
+ * actual send step. Returns true if a new row was inserted.
+ */
+export function createPendingQuestion(pq: PendingQuestion): boolean {
+  const result = getDb()
     .prepare(
-      `INSERT INTO pending_questions (question_id, session_id, message_out_id, platform_id, channel_type, thread_id, title, options_json, created_at)
+      `INSERT OR IGNORE INTO pending_questions (question_id, session_id, message_out_id, platform_id, channel_type, thread_id, title, options_json, created_at)
        VALUES (@question_id, @session_id, @message_out_id, @platform_id, @channel_type, @thread_id, @title, @options_json, @created_at)`,
     )
     .run({
@@ -114,6 +120,7 @@ export function createPendingQuestion(pq: PendingQuestion): void {
       options_json: JSON.stringify(pq.options),
       created_at: pq.created_at,
     });
+  return result.changes > 0;
 }
 
 export function getPendingQuestion(questionId: string): PendingQuestion | undefined {
@@ -131,16 +138,21 @@ export function deletePendingQuestion(questionId: string): void {
 
 // ── Pending Approvals ──
 
+/**
+ * Insert a pending approval row. Idempotent for the same reason as
+ * createPendingQuestion: delivery retries with the same approval_id must not
+ * fail on UNIQUE before the send step gets a chance to succeed.
+ */
 export function createPendingApproval(
   pa: Partial<PendingApproval> &
     Pick<
       PendingApproval,
       'approval_id' | 'request_id' | 'action' | 'payload' | 'created_at' | 'title' | 'options_json'
     >,
-): void {
-  getDb()
+): boolean {
+  const result = getDb()
     .prepare(
-      `INSERT INTO pending_approvals
+      `INSERT OR IGNORE INTO pending_approvals
          (approval_id, session_id, request_id, action, payload, created_at,
           agent_group_id, channel_type, platform_id, platform_message_id, expires_at, status,
           title, options_json)
@@ -159,6 +171,7 @@ export function createPendingApproval(
       status: 'pending',
       ...pa,
     });
+  return result.changes > 0;
 }
 
 export function getPendingApproval(approvalId: string): PendingApproval | undefined {
@@ -192,6 +205,23 @@ export function getAskQuestionRender(
   const a = getDb().prepare('SELECT title, options_json FROM pending_approvals WHERE approval_id = ?').get(id) as
     | { title: string; options_json: string }
     | undefined;
-  if (!a || !a.title) return undefined;
-  return { title: a.title, options: JSON.parse(a.options_json) };
+  if (a?.title) return { title: a.title, options: JSON.parse(a.options_json) };
+
+  // Channel-registration + unknown-sender approvals persist title/options_json
+  // the same way pending_approvals does — just SELECT and return.
+  if (hasTable(getDb(), 'pending_channel_approvals')) {
+    const c = getDb()
+      .prepare('SELECT title, options_json FROM pending_channel_approvals WHERE messaging_group_id = ?')
+      .get(id) as { title: string; options_json: string } | undefined;
+    if (c?.title) return { title: c.title, options: JSON.parse(c.options_json) };
+  }
+
+  if (hasTable(getDb(), 'pending_sender_approvals')) {
+    const s = getDb().prepare('SELECT title, options_json FROM pending_sender_approvals WHERE id = ?').get(id) as
+      | { title: string; options_json: string }
+      | undefined;
+    if (s?.title) return { title: s.title, options: JSON.parse(s.options_json) };
+  }
+
+  return undefined;
 }

src/delivery.ts

@@ -321,7 +321,7 @@ async function deliverMessage(
         questionId: content.questionId,
       });
     } else {
-      createPendingQuestion({
+      const inserted = createPendingQuestion({
         question_id: content.questionId,
         session_id: session.id,
         message_out_id: msg.id,
@@ -332,7 +332,9 @@ async function deliverMessage(
         options: normalizeOptions(rawOptions as never),
         created_at: new Date().toISOString(),
       });
-      log.info('Pending question created', { questionId: content.questionId, sessionId: session.id });
+      if (inserted) {
+        log.info('Pending question created', { questionId: content.questionId, sessionId: session.id });
+      }
     }
   }
 

src/host-core.test.ts

@@ -173,6 +173,43 @@ describe('session manager', () => {
 
     expect(getSession(session.id)!.last_active).not.toBeNull();
   });
+
+  it('should refuse path-traversal in attachment filenames', () => {
+    // Regression: attachment.name comes from untrusted senders (E2EE-protected
+    // chat platforms can't sanitize it server-side). Without the guard, a
+    // `../../../tmp/pwned` filename escapes the inbox dir and writes anywhere
+    // the host process can reach.
+    const { session } = resolveSession('ag-1', 'mg-1', null, 'shared');
+    const inboxBase = path.join(sessionDir('ag-1', session.id), 'inbox');
+    const escapeTarget = path.join('/tmp', 'nanoclaw-traversal-canary');
+    if (fs.existsSync(escapeTarget)) fs.rmSync(escapeTarget);
+
+    writeSessionMessage('ag-1', session.id, {
+      id: 'msg-attack',
+      kind: 'chat',
+      timestamp: now(),
+      content: JSON.stringify({
+        text: 'pwn',
+        attachments: [
+          {
+            type: 'document',
+            name: '../../../../../../../../tmp/nanoclaw-traversal-canary',
+            data: Buffer.from('owned').toString('base64'),
+          },
+        ],
+      }),
+    });
+
+    expect(fs.existsSync(escapeTarget)).toBe(false);
+    // The bytes should still land — under a synthesized safe name inside the
+    // inbox — so the agent doesn't lose data on a malicious filename.
+    const inboxDir = path.join(inboxBase, 'msg-attack');
+    expect(fs.existsSync(inboxDir)).toBe(true);
+    const written = fs.readdirSync(inboxDir);
+    expect(written).toHaveLength(1);
+    expect(written[0]).not.toContain('/');
+    expect(written[0]).not.toContain('..');
+  });
 });
 
 describe('router', () => {

src/host-sweep.ts

@@ -159,23 +159,31 @@ async function sweepSession(session: Session): Promise<void> {
       syncProcessingAcks(inDb, outDb);
     }
 
-    const alive = isContainerRunning(session.id);
-
-    // 2. Crashed-container cleanup: processing rows left behind get retried.
-    if (!alive && outDb) {
-      resetStuckProcessingRows(inDb, outDb, session, 'container not running');
+    // 2. Wake a container if work is due and nothing is running. Ordered
+    // before the crashed-container cleanup so a fresh container gets a chance
+    // to clean its own orphan processing_ack rows on startup (see
+    // container/agent-runner/src/db/connection.ts). Otherwise the reset path
+    // would keep bumping process_after into the future, dueCount would stay 0,
+    // and the wake would never fire.
+    const dueCount = countDueMessages(inDb);
+    if (dueCount > 0 && !isContainerRunning(session.id)) {
+      log.info('Waking container for due messages', { sessionId: session.id, count: dueCount });
+      await wakeContainer(session);
     }
 
+    const alive = isContainerRunning(session.id);
+
     // 3. Running-container SLA: absolute ceiling + per-claim stuck rules.
     if (alive && outDb) {
       enforceRunningContainerSla(inDb, outDb, session, agentGroup.id);
     }
 
-    // 4. Wake a container if new work is due and nothing is running.
-    const dueCount = countDueMessages(inDb);
-    if (dueCount > 0 && !isContainerRunning(session.id)) {
-      log.info('Waking container for due messages', { sessionId: session.id, count: dueCount });
-      await wakeContainer(session);
+    // 4. Crashed-container cleanup: processing rows left behind get retried.
+    // Only fires when wake in step 2 didn't pick up the work (no due messages,
+    // or wake failed). resetStuckProcessingRows itself is idempotent — it
+    // skips messages already scheduled for a future retry.
+    if (!alive && outDb) {
+      resetStuckProcessingRows(inDb, outDb, session, 'container not running');
     }
 
     // 5. Recurrence fanout for completed recurring tasks.
@@ -246,10 +254,16 @@ function resetStuckProcessingRows(
   reason: string,
 ): void {
   const claims = getProcessingClaims(outDb);
+  const now = Date.now();
   for (const { message_id } of claims) {
     const msg = getMessageForRetry(inDb, message_id, 'pending');
     if (!msg) continue;
 
+    // Already rescheduled for a future retry — don't bump tries again. The
+    // wake path (sweep step 2) will fire when process_after elapses and a
+    // fresh container will clean the orphan claim on startup.
+    if (msg.processAfter && Date.parse(msg.processAfter) > now) continue;
+
     if (msg.tries >= MAX_TRIES) {
       markMessageFailed(inDb, msg.id);
       log.warn('Message marked as failed after max retries', {

src/index.ts

@@ -85,6 +85,7 @@ async function main(): Promise<void> {
             content: JSON.stringify(message.content),
             timestamp: message.timestamp,
             isMention: message.isMention,
+            isGroup: message.isGroup,
           },
         }).catch((err) => {
           log.error('Failed to route inbound message', { channelType: adapter.channelType, err });

src/modules/agent-to-agent/agent-route.test.ts

@@ -0,0 +1,46 @@
+import { describe, expect, it } from 'vitest';
+
+import { isSafeAttachmentName } from './agent-route.js';
+
+/**
+ * `forwardAttachedFiles` has a filesystem side that's awkward to unit-test
+ * without mocking DATA_DIR. The guarantee worth pinning is that the
+ * filename validator rejects everything that could escape the inbox dir —
+ * `forwardAttachedFiles` runs this guard before any I/O, so traversal is
+ * impossible as long as this matrix holds.
+ */
+describe('isSafeAttachmentName', () => {
+  it('accepts plain filenames', () => {
+    expect(isSafeAttachmentName('baby-duck.png')).toBe(true);
+    expect(isSafeAttachmentName('file with spaces.pdf')).toBe(true);
+    expect(isSafeAttachmentName('report.v2.docx')).toBe(true);
+    expect(isSafeAttachmentName('.hidden')).toBe(true); // leading dot is fine, just not `.` / `..`
+  });
+
+  it('rejects empty / sentinel values', () => {
+    expect(isSafeAttachmentName('')).toBe(false);
+    expect(isSafeAttachmentName('.')).toBe(false);
+    expect(isSafeAttachmentName('..')).toBe(false);
+  });
+
+  it('rejects path separators', () => {
+    expect(isSafeAttachmentName('../evil.png')).toBe(false);
+    expect(isSafeAttachmentName('/etc/passwd')).toBe(false);
+    expect(isSafeAttachmentName('nested/file.txt')).toBe(false);
+    expect(isSafeAttachmentName('windows\\path.exe')).toBe(false);
+  });
+
+  it('rejects NUL bytes', () => {
+    expect(isSafeAttachmentName('clean\0.png')).toBe(false);
+  });
+
+  it('rejects anything path.basename would strip', () => {
+    expect(isSafeAttachmentName('a/b')).toBe(false);
+    expect(isSafeAttachmentName('./thing')).toBe(false);
+  });
+
+  it('rejects non-string input', () => {
+    expect(isSafeAttachmentName(null as unknown as string)).toBe(false);
+    expect(isSafeAttachmentName(undefined as unknown as string)).toBe(false);
+  });
+});

src/modules/agent-to-agent/agent-route.ts

@@ -3,9 +3,13 @@
  *
  * Outbound messages with `channel_type === 'agent'` target another agent
  * group rather than a channel. Permission is enforced via `agent_destinations` —
- * the source agent must have a row for the target. Content is copied verbatim;
- * the target's formatter looks up the source agent in its own local map to
- * display a name.
+ * the source agent must have a row for the target. Content is copied into the
+ * target's inbound DB; if the source message had `files` (from `send_file`),
+ * the actual bytes are copied from the source's outbox into the target's
+ * `inbox/<a2a-msg-id>/` directory and surfaced to the target agent as
+ * `attachments` (existing formatter convention — see formatter.ts:230).
+ * The target agent can then forward the file onward via its own `send_file`
+ * call using the absolute `/workspace/inbox/<a2a-msg-id>/<filename>` path.
  *
  * Self-messages are always allowed (used for system notes injected back into
  * an agent's own session, e.g. post-approval follow-up prompts).
@@ -14,14 +18,85 @@
  * `channel_type === 'agent'` check. When the module is absent the check in
  * core throws with a "module not installed" message so retry → mark failed.
  */
+import fs from 'fs';
+import path from 'path';
+
+import { isSafeAttachmentName } from '../../attachment-safety.js';
 import { getAgentGroup } from '../../db/agent-groups.js';
 import { getSession } from '../../db/sessions.js';
 import { wakeContainer } from '../../container-runner.js';
 import { log } from '../../log.js';
-import { resolveSession, writeSessionMessage } from '../../session-manager.js';
+import { resolveSession, sessionDir, writeSessionMessage } from '../../session-manager.js';
 import type { Session } from '../../types.js';
 import { hasDestination } from './db/agent-destinations.js';
 
+export { isSafeAttachmentName };
+
+export interface ForwardedAttachment {
+  name: string;
+  filename: string;
+  type: 'file';
+  localPath: string;
+}
+
+/**
+ * Copy file attachments from the source agent's outbox into the target
+ * agent's inbox. Returns attachments using the formatter's existing
+ * `{name, type, localPath}` convention — target agent reads `localPath`
+ * as relative to `/workspace/`, matching how channel-inbound attachments
+ * are surfaced today.
+ *
+ * Missing source files and unsafe (path-traversal) filenames are skipped
+ * with a warning rather than failing the whole route — a bad filename
+ * reference shouldn't kill the accompanying text.
+ */
+export function forwardAttachedFiles(
+  source: { agentGroupId: string; sessionId: string; messageId: string; filenames: string[] },
+  target: { agentGroupId: string; sessionId: string; messageId: string },
+): ForwardedAttachment[] {
+  if (source.filenames.length === 0) return [];
+
+  const sourceDir = path.join(sessionDir(source.agentGroupId, source.sessionId), 'outbox', source.messageId);
+  if (!fs.existsSync(sourceDir)) {
+    log.warn('agent-route: source outbox dir missing, no files forwarded', {
+      sourceMsgId: source.messageId,
+      sourceDir,
+    });
+    return [];
+  }
+
+  const targetInboxDir = path.join(sessionDir(target.agentGroupId, target.sessionId), 'inbox', target.messageId);
+  fs.mkdirSync(targetInboxDir, { recursive: true });
+
+  const attachments: ForwardedAttachment[] = [];
+  for (const filename of source.filenames) {
+    if (!isSafeAttachmentName(filename)) {
+      log.warn('agent-route: rejecting unsafe attachment filename (path traversal attempt?)', {
+        sourceMsgId: source.messageId,
+        filename,
+      });
+      continue;
+    }
+    const src = path.join(sourceDir, filename);
+    if (!fs.existsSync(src)) {
+      log.warn('agent-route: referenced file missing in source outbox, skipped', {
+        sourceMsgId: source.messageId,
+        filename,
+      });
+      continue;
+    }
+    const dst = path.join(targetInboxDir, filename);
+    fs.copyFileSync(src, dst);
+    attachments.push({
+      name: filename,
+      filename,
+      type: 'file',
+      localPath: `inbox/${target.messageId}/${filename}`,
+    });
+  }
+  return attachments;
+}
+
 export interface RoutableAgentMessage {
   id: string;
   platform_id: string | null;
@@ -45,20 +120,87 @@ export async function routeAgentMessage(msg: RoutableAgentMessage, session: Sess
     throw new Error(`target agent group ${targetAgentGroupId} not found for message ${msg.id}`);
   }
   const { session: targetSession } = resolveSession(targetAgentGroupId, null, null, 'agent-shared');
+  const a2aMsgId = `a2a-${Date.now()}-${Math.random().toString(36).slice(2, 8)}`;
+
+  // If the source message references files (via `send_file`), forward the
+  // bytes from the source's outbox into the target's inbox so the target
+  // agent can actually see and re-send them. Without this, agent-to-agent
+  // file attachments look like they arrive but the target has no way to
+  // read the bytes — they live in a session dir it doesn't mount.
+  const forwardedContent = forwardFileAttachments(msg, a2aMsgId, session, targetAgentGroupId, targetSession.id);
+
   writeSessionMessage(targetAgentGroupId, targetSession.id, {
-    id: `a2a-${Date.now()}-${Math.random().toString(36).slice(2, 8)}`,
+    id: a2aMsgId,
     kind: 'chat',
     timestamp: new Date().toISOString(),
     platformId: session.agent_group_id,
     channelType: 'agent',
     threadId: null,
-    content: msg.content,
+    content: forwardedContent,
   });
   log.info('Agent message routed', {
     from: session.agent_group_id,
     to: targetAgentGroupId,
     targetSession: targetSession.id,
+    a2aMsgId,
+    forwardedFileCount: countForwardedFiles(forwardedContent),
   });
   const fresh = getSession(targetSession.id);
   if (fresh) await wakeContainer(fresh);
 }
+
+/**
+ * Parse source content, copy any referenced `files` from source outbox to
+ * target inbox, and return a JSON string with an `attachments` array added
+ * (formatter.ts:223 already knows how to render this shape).
+ *
+ * If the source content isn't JSON or has no files, returns the original
+ * content string unchanged — this is safe to call on every route.
+ */
+function forwardFileAttachments(
+  msg: RoutableAgentMessage,
+  a2aMsgId: string,
+  sourceSession: Session,
+  targetAgentGroupId: string,
+  targetSessionId: string,
+): string {
+  let parsed: Record<string, unknown>;
+  try {
+    parsed = JSON.parse(msg.content);
+  } catch {
+    return msg.content;
+  }
+  const files = parsed.files as unknown;
+  if (!Array.isArray(files) || files.length === 0) return msg.content;
+  const filenames = files.filter((f): f is string => typeof f === 'string');
+  if (filenames.length === 0) return msg.content;
+
+  const attachments = forwardAttachedFiles(
+    {
+      agentGroupId: sourceSession.agent_group_id,
+      sessionId: sourceSession.id,
+      messageId: msg.id,
+      filenames,
+    },
+    {
+      agentGroupId: targetAgentGroupId,
+      sessionId: targetSessionId,
+      messageId: a2aMsgId,
+    },
+  );
+
+  // Merge into any existing `attachments` (unlikely in a2a context but safe).
+  const existing = Array.isArray(parsed.attachments) ? (parsed.attachments as Record<string, unknown>[]) : [];
+  parsed.attachments = [...existing, ...attachments];
+
+  return JSON.stringify(parsed);
+}
+
+function countForwardedFiles(contentStr: string): number {
+  try {
+    const parsed = JSON.parse(contentStr);
+    return Array.isArray(parsed.attachments) ? parsed.attachments.length : 0;
+  } catch {
+    return 0;
+  }
+}

src/modules/permissions/channel-approval.ts

@@ -101,13 +101,26 @@ export async function requestChannelApproval(input: RequestChannelApprovalInput)
     return;
   }
 
-  const originName = originMg?.name ?? originMg?.platform_id ?? 'an unfamiliar chat';
-  const isGroup = originMg?.is_group === 1;
+  const isGroup = event.message?.isGroup ?? originMg?.is_group === 1;
+
+  // Extract sender name from the event content for a human-readable card.
+  let senderName: string | undefined;
+  try {
+    const parsed = JSON.parse(event.message.content) as Record<string, unknown>;
+    senderName = (parsed.senderName ?? parsed.sender) as string | undefined;
+  } catch {
+    // non-critical — fall through to generic wording
+  }
 
   const title = isGroup ? '📣 Bot mentioned in new chat' : '💬 New direct message';
   const question = isGroup
-    ? `Your agent was mentioned in ${originName} on ${originChannelType}. Wire it to ${target.name} and let it engage?`
-    : `Someone DM'd your agent on ${originChannelType} (${originName}). Wire it to ${target.name} and let it respond?`;
+    ? senderName
+      ? `${senderName} mentioned your agent in a ${originChannelType} channel. Wire it to ${target.name} and let it engage?`
+      : `Your agent was mentioned in a ${originChannelType} channel. Wire it to ${target.name} and let it engage?`
+    : senderName
+      ? `${senderName} DM'd your agent on ${originChannelType}. Wire it to ${target.name} and let it respond?`
+      : `Someone DM'd your agent on ${originChannelType}. Wire it to ${target.name} and let it respond?`;
+  const options = normalizeOptions(APPROVAL_OPTIONS);
 
   createPendingChannelApproval({
     messaging_group_id: messagingGroupId,
@@ -115,6 +128,8 @@ export async function requestChannelApproval(input: RequestChannelApprovalInput)
     original_message: JSON.stringify(event),
     approver_user_id: delivery.userId,
     created_at: new Date().toISOString(),
+    title,
+    options_json: JSON.stringify(options),
   });
 
   const adapter = getDeliveryAdapter();
@@ -139,7 +154,7 @@ export async function requestChannelApproval(input: RequestChannelApprovalInput)
         questionId: messagingGroupId,
         title,
         question,
-        options: normalizeOptions(APPROVAL_OPTIONS),
+        options,
       }),
     );
     log.info('Channel registration card delivered', {

src/modules/permissions/db/pending-channel-approvals.ts

@@ -17,6 +17,10 @@ export interface PendingChannelApproval {
   original_message: string;
   approver_user_id: string;
   created_at: string;
+  /** Card title shown at creation and re-used by getAskQuestionRender on click. */
+  title: string;
+  /** Normalized options (JSON-encoded NormalizedOption[]) — same shape persisted on pending_approvals. */
+  options_json: string;
 }
 
 export function createPendingChannelApproval(row: PendingChannelApproval): void {
@@ -24,11 +28,11 @@ export function createPendingChannelApproval(row: PendingChannelApproval): void
     .prepare(
       `INSERT INTO pending_channel_approvals (
          messaging_group_id, agent_group_id, original_message,
-         approver_user_id, created_at
+         approver_user_id, created_at, title, options_json
        )
        VALUES (
          @messaging_group_id, @agent_group_id, @original_message,
-         @approver_user_id, @created_at
+         @approver_user_id, @created_at, @title, @options_json
        )`,
     )
     .run(row);

src/modules/permissions/db/pending-sender-approvals.ts

@@ -19,6 +19,10 @@ export interface PendingSenderApproval {
   original_message: string;
   approver_user_id: string;
   created_at: string;
+  /** Card title shown at creation and re-used by getAskQuestionRender on click. */
+  title: string;
+  /** Normalized options (JSON-encoded NormalizedOption[]) — same shape persisted on pending_approvals. */
+  options_json: string;
 }
 
 export function createPendingSenderApproval(row: PendingSenderApproval): void {
@@ -26,11 +30,13 @@ export function createPendingSenderApproval(row: PendingSenderApproval): void {
     .prepare(
       `INSERT INTO pending_sender_approvals (
          id, messaging_group_id, agent_group_id, sender_identity,
-         sender_name, original_message, approver_user_id, created_at
+         sender_name, original_message, approver_user_id, created_at,
+         title, options_json
        )
        VALUES (
          @id, @messaging_group_id, @agent_group_id, @sender_identity,
-         @sender_name, @original_message, @approver_user_id, @created_at
+         @sender_name, @original_message, @approver_user_id, @created_at,
+         @title, @options_json
        )`,
     )
     .run(row);

src/modules/permissions/sender-approval.ts

@@ -88,10 +88,11 @@ export async function requestSenderApproval(input: RequestSenderApprovalInput):
 
   const approvalId = generateId();
   const senderDisplay = senderName && senderName.length > 0 ? senderName : senderIdentity;
-  const originName = originMg?.name ?? originMg?.platform_id ?? 'an unfamiliar chat';
+  const originName = originMg?.name ?? `a ${originChannelType} channel`;
 
   const title = '👤 New sender';
   const question = `${senderDisplay} wants to talk to your agent in ${originName}. Allow?`;
+  const options = normalizeOptions(APPROVAL_OPTIONS);
 
   createPendingSenderApproval({
     id: approvalId,
@@ -102,6 +103,8 @@ export async function requestSenderApproval(input: RequestSenderApprovalInput):
     original_message: JSON.stringify(event),
     approver_user_id: target.userId,
     created_at: new Date().toISOString(),
+    title,
+    options_json: JSON.stringify(options),
   });
 
   const adapter = getDeliveryAdapter();
@@ -126,7 +129,7 @@ export async function requestSenderApproval(input: RequestSenderApprovalInput):
         questionId: approvalId,
         title,
         question,
-        options: APPROVAL_OPTIONS,
+        options,
       }),
     );
     log.info('Unknown-sender approval card delivered', {

src/platform-id.ts

@@ -0,0 +1,23 @@
+/**
+ * Determine whether a platform ID needs a channel-type prefix.
+ *
+ * Chat SDK adapters (Telegram, Discord, Slack, Teams, etc.) namespace their
+ * platform IDs with a channel prefix: "telegram:123456", "discord:guild:chan".
+ * The router stores channel_type and platform_id in separate columns, but
+ * Chat SDK adapters send the prefixed form as the platform_id — so any code
+ * that writes messaging_groups rows must produce the same shape the adapter
+ * will later emit as event.platformId, or router lookups miss and messages
+ * get silently dropped.
+ *
+ * Native adapters (Signal, WhatsApp, iMessage) use their own ID formats and
+ * send them as-is — no channel prefix. WhatsApp/iMessage emit JIDs/emails
+ * containing '@'. Signal emits raw phone numbers ('+15551234567') for DMs
+ * and 'group:<id>' for group chats. Prefixing any of these would cause a
+ * mismatch with what the adapter later emits.
+ */
+export function namespacedPlatformId(channel: string, raw: string): string {
+  if (raw.startsWith(`${channel}:`)) return raw;
+  if (raw.includes('@')) return raw;
+  if (raw.startsWith('+') || raw.startsWith('group:')) return raw;
+  return `${channel}:${raw}`;
+}

src/providers/claude.ts

@@ -0,0 +1,28 @@
+/**
+ * Claude provider container config — only registered when the user has
+ * configured a custom Anthropic-compatible endpoint via setup. Setup
+ * appends `import './claude.js'` to providers/index.ts at that point;
+ * standard installs hitting api.anthropic.com don't need this file
+ * loaded.
+ *
+ * The real auth token never enters the container. Setup creates an
+ * OneCLI generic secret (host-pattern = base URL hostname, header-name
+ * = Authorization, value-format = "Bearer {value}") so the proxy
+ * rewrites the Authorization header on the wire. The container only
+ * needs:
+ *   - ANTHROPIC_BASE_URL — so the SDK knows where to call
+ *   - ANTHROPIC_AUTH_TOKEN=placeholder — so the SDK adds an
+ *     Authorization: Bearer header for OneCLI to overwrite
+ */
+import { readEnvFile } from '../env.js';
+import { registerProviderContainerConfig } from './provider-container-registry.js';
+
+registerProviderContainerConfig('claude', () => {
+  const dotenv = readEnvFile(['ANTHROPIC_BASE_URL']);
+  const env: Record<string, string> = {};
+  if (dotenv.ANTHROPIC_BASE_URL) {
+    env.ANTHROPIC_BASE_URL = dotenv.ANTHROPIC_BASE_URL;
+    env.ANTHROPIC_AUTH_TOKEN = 'placeholder';
+  }
+  return { env };
+});

src/providers/codex-agents-md.test.ts

@@ -0,0 +1,112 @@
+/**
+ * The 32KB Codex project-doc cap must DEGRADE, never throw: composeGroupAgentsMd
+ * runs inside the provider contribution at every spawn, and a throw there rides
+ * wakeContainer's transient-retry contract — host-sweep respawns every 60s
+ * forever and the group goes silently dark (a permanent condition disguised as
+ * a transient one). Oversized docs drop their largest optional instruction
+ * sections, keep the core contract, and say so in the doc.
+ */
+import fs from 'fs';
+import os from 'os';
+import path from 'path';
+import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
+
+vi.mock('../config.js', async (importOriginal) => ({
+  ...(await importOriginal<typeof import('../config.js')>()),
+  DATA_DIR: '/tmp/nanoclaw-agents-md-test/data',
+}));
+
+import { composeGroupAgentsMd, CODEX_PROJECT_DOC_MAX_BYTES } from './codex-agents-md.js';
+import { closeDb, createAgentGroup, initTestDb, runMigrations } from '../db/index.js';
+import { ensureContainerConfig, updateContainerConfigJson } from '../db/container-configs.js';
+import type { AgentGroup } from '../types.js';
+
+const TEST_ROOT = '/tmp/nanoclaw-agents-md-test';
+
+function group(folder: string): AgentGroup {
+  return {
+    id: `ag-${folder}`,
+    name: folder,
+    folder,
+    agent_provider: null,
+    created_at: new Date().toISOString(),
+  } as AgentGroup;
+}
+
+describe('composeGroupAgentsMd cap handling', () => {
+  beforeEach(() => {
+    if (fs.existsSync(TEST_ROOT)) fs.rmSync(TEST_ROOT, { recursive: true });
+    fs.mkdirSync(path.join(TEST_ROOT, 'data'), { recursive: true });
+    const db = initTestDb();
+    runMigrations(db);
+  });
+
+  afterEach(() => {
+    closeDb();
+    if (fs.existsSync(TEST_ROOT)) fs.rmSync(TEST_ROOT, { recursive: true });
+  });
+
+  it('writes the doc untouched when under the cap', () => {
+    const g = group('small');
+    createAgentGroup(g);
+    ensureContainerConfig(g.id);
+    const groupDir = fs.mkdtempSync(path.join(os.tmpdir(), 'agents-md-'));
+    try {
+      composeGroupAgentsMd(g, groupDir);
+      const doc = fs.readFileSync(path.join(groupDir, 'AGENTS.md'), 'utf-8');
+      expect(doc).not.toContain('Omitted for size');
+      // Agent-authored skills must be told their persistent home — without
+      // this, authored skills land on ephemeral container paths and vanish.
+      expect(doc).toContain('/workspace/agent/skills');
+      expect(Buffer.byteLength(doc, 'utf-8')).toBeLessThanOrEqual(CODEX_PROJECT_DOC_MAX_BYTES);
+    } finally {
+      fs.rmSync(groupDir, { recursive: true, force: true });
+    }
+  });
+
+  it('inlines the memory index so recall does not depend on a file read', () => {
+    const g = group('with-memory');
+    createAgentGroup(g);
+    ensureContainerConfig(g.id);
+    const groupDir = fs.mkdtempSync(path.join(os.tmpdir(), 'agents-md-'));
+    try {
+      fs.mkdirSync(path.join(groupDir, 'memory'), { recursive: true });
+      fs.writeFileSync(
+        path.join(groupDir, 'memory', 'index.md'),
+        '# Memory Index\n- [People](memories/people/) - notes about people and their preferences\n',
+      );
+
+      composeGroupAgentsMd(g, groupDir);
+
+      const doc = fs.readFileSync(path.join(groupDir, 'AGENTS.md'), 'utf-8');
+      expect(doc).toContain('Current memory index');
+      expect(doc).toContain('notes about people and their preferences');
+    } finally {
+      fs.rmSync(groupDir, { recursive: true, force: true });
+    }
+  });
+
+  it('degrades instead of throwing when MCP instructions push the doc over the cap', () => {
+    const g = group('oversized');
+    createAgentGroup(g);
+    ensureContainerConfig(g.id);
+    updateContainerConfigJson(g.id, 'mcp_servers', {
+      bloated: { command: 'x', instructions: 'y'.repeat(CODEX_PROJECT_DOC_MAX_BYTES + 1024) },
+      lean: { command: 'x', instructions: 'short and useful' },
+    });
+    const groupDir = fs.mkdtempSync(path.join(os.tmpdir(), 'agents-md-'));
+    try {
+      composeGroupAgentsMd(g, groupDir); // must not throw
+
+      const doc = fs.readFileSync(path.join(groupDir, 'AGENTS.md'), 'utf-8');
+      expect(Buffer.byteLength(doc, 'utf-8')).toBeLessThanOrEqual(CODEX_PROJECT_DOC_MAX_BYTES);
+      // Largest optional section dropped, named in the doc; the rest survive.
+      expect(doc).toContain('Omitted for size');
+      expect(doc).toContain('MCP Server: bloated');
+      expect(doc).toContain('short and useful');
+      expect(doc).toContain('Memory System');
+    } finally {
+      fs.rmSync(groupDir, { recursive: true, force: true });
+    }
+  });
+});

src/providers/codex-agents-md.ts

@@ -0,0 +1,188 @@
+/**
+ * AGENTS.md composition for codex agent groups — codex-owned payload code.
+ *
+ * AGENTS.md is Codex's project doc (its CLAUDE.md equivalent). Composed fresh
+ * on every spawn by the codex provider contribution (see ./codex.ts) from:
+ *   - the shared base (`container/AGENTS.md`)
+ *   - a pointer to the runner-scaffolded memory system (created container-side
+ *     at boot via the `usesMemoryScaffold` capability — nothing is written here)
+ *   - a pointer to codex-native skills under `.agents/skills`
+ *   - each enabled NanoClaw module's `*.instructions.md` fragment
+ *   - MCP server `instructions` from container.json
+ *
+ * Codex hard-caps project-doc loading (`project_doc_max_bytes`, mirrored in
+ * the container provider's config.toml writer) — compose fails loudly rather
+ * than letting Codex truncate silently.
+ */
+import fs from 'fs';
+import path from 'path';
+
+import type { McpServerConfig } from '../container-config.js';
+import { getContainerConfig } from '../db/container-configs.js';
+import { log } from '../log.js';
+import type { AgentGroup } from '../types.js';
+
+export const CODEX_PROJECT_DOC_MAX_BYTES = 32 * 1024;
+export const CODEX_PROJECT_DOC_WARN_BYTES = 28 * 1024;
+
+const HEADER = '<!-- Composed at spawn. Do not edit. Edit memory/system/definition.md for memory behavior. -->';
+const MCP_TOOLS_HOST_SUBPATH = path.join('container', 'agent-runner', 'src', 'mcp-tools');
+
+const MEMORY_POINTER = [
+  'Editable memory-system definition: `/workspace/agent/memory/system/definition.md`.',
+  'Top memory index: `/workspace/agent/memory/index.md`.',
+  'Read the definition and index, then use memories, data, and conversation archives when relevant.',
+  'Stored user preferences are binding: before your first reply in a session, check the index below and read any memory file relevant to the user or the request, and apply it without being asked.',
+  'Do not use `AGENTS.local.md` or `AGENTS.override.md` for memory.',
+].join('\n\n');
+
+/**
+ * Inline the group's current memory index into the composed doc. Recall must
+ * not depend on the model choosing to read a file before its first reply —
+ * with the map already in the system prompt, applying a stored preference is
+ * one hop (read the relevant memory file), not three. The index is small
+ * (hundreds of bytes); the 32KB fit logic above bounds the worst case.
+ */
+function memoryIndexInline(groupDir: string): string {
+  const indexPath = path.join(groupDir, 'memory', 'index.md');
+  if (!fs.existsSync(indexPath)) return '';
+  const content = fs.readFileSync(indexPath, 'utf-8').trim();
+  if (!content) return '';
+  return ['Current memory index (paths relative to `/workspace/agent/memory/`):', content].join('\n\n');
+}
+
+const NATIVE_RUNTIME_SKILLS_POINTER = [
+  'Selected NanoClaw runtime skills are available as Codex-native skills at `/workspace/agent/.agents/skills`.',
+  'Each skill directory contains a `SKILL.md` with its trigger description plus any supporting files, and points to the read-only shared skill source under `/app/skills`.',
+  'Use skill discovery to load these skills only when their descriptions match the task. Full skill instructions live in the skill directories, not in `AGENTS.md`.',
+  'Skills YOU author or install yourself go in `/workspace/agent/skills/<name>/SKILL.md` — persistent, provider-neutral (they load under any agent provider this group runs on), and yours to write and update over time. They are linked into `$CODEX_HOME/skills` automatically at boot. Never write skills anywhere else: paths outside your workspace and `$CODEX_HOME` are ephemeral.',
+].join('\n\n');
+
+interface AgentsMdSection {
+  name: string;
+  content: string;
+}
+
+export function composeGroupAgentsMd(group: AgentGroup, groupDir: string): void {
+  if (!fs.existsSync(groupDir)) fs.mkdirSync(groupDir, { recursive: true });
+
+  const configRow = getContainerConfig(group.id);
+  const mcpServers: Record<string, McpServerConfig> = configRow
+    ? (JSON.parse(configRow.mcp_servers) as Record<string, McpServerConfig>)
+    : {};
+
+  const sections: AgentsMdSection[] = [{ name: 'header', content: HEADER }];
+  const pushSection = (name: string, ...content: string[]): void => {
+    const body = content
+      .map((part) => part.trim())
+      .filter(Boolean)
+      .join('\n\n');
+    if (body) sections.push({ name, content: `# ${name}\n\n${body}` });
+  };
+
+  const sharedBase = path.join(process.cwd(), 'container', 'AGENTS.md');
+  if (fs.existsSync(sharedBase)) {
+    pushSection('NanoClaw Runtime Contract', fs.readFileSync(sharedBase, 'utf-8'));
+  }
+
+  pushSection('Memory System', MEMORY_POINTER, memoryIndexInline(groupDir));
+  pushSection('Native Runtime Skills', NATIVE_RUNTIME_SKILLS_POINTER);
+
+  const cliDisabled = configRow?.cli_scope === 'disabled';
+  const mcpToolsHostDir = path.join(process.cwd(), MCP_TOOLS_HOST_SUBPATH);
+  if (fs.existsSync(mcpToolsHostDir)) {
+    for (const entry of fs.readdirSync(mcpToolsHostDir).sort()) {
+      const match = entry.match(/^(.+)\.instructions\.md$/);
+      if (!match) continue;
+      const moduleName = match[1];
+      if (moduleName === 'cli' && cliDisabled) continue;
+      pushSection(`NanoClaw Module: ${moduleName}`, fs.readFileSync(path.join(mcpToolsHostDir, entry), 'utf-8'));
+    }
+  }
+
+  for (const [name, mcp] of Object.entries(mcpServers)) {
+    if (mcp.instructions) {
+      pushSection(`MCP Server: ${name}`, mcp.instructions);
+    }
+  }
+
+  const content = fitAgentsMdToCap(group, sections);
+  writeAtomic(path.join(groupDir, 'AGENTS.md'), content);
+}
+
+function renderAgentsMd(sections: AgentsMdSection[]): string {
+  return (
+    sections
+      .map((section) => section.content.trim())
+      .filter(Boolean)
+      .join('\n\n') + '\n'
+  );
+}
+
+/**
+ * Fit the doc under Codex's 32KB project-doc cap by DEGRADING, never
+ * throwing: a per-spawn throw rides wakeContainer's transient-retry contract
+ * — host-sweep respawns every 60s forever and the group goes silently dark.
+ * Instead, drop the largest optional instruction sections (per-module and
+ * per-MCP-server) until the doc fits, log what was dropped at error level,
+ * and tell the agent in the doc itself. The core contract (header, runtime
+ * contract, memory, skills pointer) is never dropped.
+ */
+function fitAgentsMdToCap(group: AgentGroup, sections: AgentsMdSection[]): string {
+  const sectionBytes = (): { section: string; bytes: number }[] =>
+    sections.map((section) => ({ section: section.name, bytes: Buffer.byteLength(section.content, 'utf-8') }));
+
+  const isDroppable = (s: AgentsMdSection): boolean =>
+    s.name.startsWith('MCP Server: ') || s.name.startsWith('NanoClaw Module: ');
+
+  const dropped: string[] = [];
+  const render = (): string => {
+    const parts = [...sections];
+    if (dropped.length > 0) {
+      parts.push({
+        name: 'omitted',
+        content:
+          `# Omitted for size\n\nThese instruction sections were omitted to fit Codex's project-doc cap: ` +
+          `${dropped.join(', ')}. Their tools still work; consult each tool's own description.`,
+      });
+    }
+    return renderAgentsMd(parts);
+  };
+
+  let content = render();
+  while (Buffer.byteLength(content, 'utf-8') > CODEX_PROJECT_DOC_MAX_BYTES) {
+    const candidates = sections
+      .filter(isDroppable)
+      .sort((a, b) => Buffer.byteLength(b.content, 'utf-8') - Buffer.byteLength(a.content, 'utf-8'));
+    if (candidates.length === 0) break; // only core left — write oversized rather than brick the group
+    sections.splice(sections.indexOf(candidates[0]), 1);
+    dropped.push(candidates[0].name);
+    content = render();
+  }
+
+  const bytes = Buffer.byteLength(content, 'utf-8');
+  if (dropped.length > 0) {
+    log.error('AGENTS.md exceeded Codex project-doc cap — dropped largest instruction sections', {
+      group: group.name,
+      bytes,
+      maxBytes: CODEX_PROJECT_DOC_MAX_BYTES,
+      dropped,
+      sections: sectionBytes(),
+    });
+  } else if (bytes >= CODEX_PROJECT_DOC_WARN_BYTES) {
+    log.warn('AGENTS.md is near Codex project-doc cap', {
+      group: group.name,
+      bytes,
+      warnBytes: CODEX_PROJECT_DOC_WARN_BYTES,
+      maxBytes: CODEX_PROJECT_DOC_MAX_BYTES,
+      sections: sectionBytes(),
+    });
+  }
+  return content;
+}
+
+function writeAtomic(filePath: string, content: string): void {
+  const tmp = `${filePath}.tmp-${process.pid}`;
+  fs.writeFileSync(tmp, content);
+  fs.renameSync(tmp, filePath);
+}

src/providers/codex-host-contribution.test.ts

@@ -0,0 +1,98 @@
+/**
+ * In-process seam test for the codex HOST contribution's runtime consumption
+ * of core (the "consumes core" leg the skill guidelines require): drive the
+ * REAL registered contribution — via the real barrel and registry, never by
+ * importing codex.ts's internals — against a real test DB and a temp
+ * GROUPS_DIR/DATA_DIR, then hand its result to the real buildMounts.
+ *
+ * This is what catches core drift that typecheck can't: the
+ * DATA_DIR/v2-sessions/<id>/.codex-shared session layout, the
+ * getAgentGroup/getContainerConfig reads, the mcp_servers JSON shape consumed
+ * by composeGroupAgentsMd, and the mount set buildMounts assembles for a
+ * surfaces-providing provider. (codex-registration.test.ts only guards that
+ * the name is registered; provider-surfaces.test.ts drives a FAKE provider to
+ * test the seam itself.)
+ */
+import fs from 'fs';
+import path from 'path';
+import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
+
+const TEST_ROOT = '/tmp/nanoclaw-codex-host-contribution-test';
+const DATA_DIR = path.join(TEST_ROOT, 'data');
+const GROUPS_DIR = path.join(TEST_ROOT, 'groups');
+
+vi.mock('../config.js', async (importOriginal) => ({
+  ...(await importOriginal<typeof import('../config.js')>()),
+  DATA_DIR: '/tmp/nanoclaw-codex-host-contribution-test/data',
+  GROUPS_DIR: '/tmp/nanoclaw-codex-host-contribution-test/groups',
+}));
+
+import { buildMounts } from '../container-runner.js';
+import { closeDb, createAgentGroup, initTestDb, runMigrations } from '../db/index.js';
+import { ensureContainerConfig, updateContainerConfigJson } from '../db/container-configs.js';
+import { getProviderContainerConfig } from './provider-container-registry.js';
+import './index.js'; // the real host provider barrel
+import type { ContainerConfig } from '../container-config.js';
+import type { AgentGroup, Session } from '../types.js';
+
+function group(id: string, folder: string): AgentGroup {
+  return { id, name: folder, folder, agent_provider: null, created_at: new Date().toISOString() } as AgentGroup;
+}
+
+describe('codex host contribution against real core', () => {
+  beforeEach(() => {
+    fs.rmSync(TEST_ROOT, { recursive: true, force: true });
+    fs.mkdirSync(DATA_DIR, { recursive: true });
+    fs.mkdirSync(GROUPS_DIR, { recursive: true });
+    runMigrations(initTestDb());
+  });
+
+  afterEach(() => {
+    closeDb();
+    fs.rmSync(TEST_ROOT, { recursive: true, force: true });
+  });
+
+  it('creates the per-group state dir, composes AGENTS.md from the real config row, and mounts both', () => {
+    const ag = group('ag-codex', 'codex-group');
+    createAgentGroup(ag);
+    ensureContainerConfig(ag.id);
+    updateContainerConfigJson(ag.id, 'mcp_servers', {
+      tooling: { command: 'x', instructions: 'use the tooling server for builds' },
+    });
+    const groupDir = path.join(GROUPS_DIR, ag.folder);
+
+    const contributionFn = getProviderContainerConfig('codex');
+    expect(contributionFn).toBeDefined();
+    const contribution = contributionFn!({
+      sessionDir: path.join(DATA_DIR, 'v2-sessions', ag.id, 'session-1'),
+      agentGroupId: ag.id,
+      groupDir,
+      selectedSkills: [],
+      hostEnv: process.env,
+    });
+
+    // Per-group codex state dir exists and is mounted RW at ~/.codex.
+    const codexShared = path.join(DATA_DIR, 'v2-sessions', ag.id, '.codex-shared');
+    expect(fs.existsSync(codexShared)).toBe(true);
+    // OneCLI's auth-stub mountpoint is pre-created — on macOS Docker can't
+    // create a missing file mountpoint inside a virtiofs dir mount (exit 125
+    // on first spawn). Red here = the pre-create line was dropped.
+    expect(fs.existsSync(path.join(codexShared, 'auth.json'))).toBe(true);
+    const codexMount = contribution.mounts?.find((m) => m.containerPath === '/home/node/.codex');
+    expect(codexMount).toMatchObject({ hostPath: codexShared, readonly: false });
+
+    // AGENTS.md composed from the real DB row — MCP instructions included.
+    const agentsMd = fs.readFileSync(path.join(groupDir, 'AGENTS.md'), 'utf-8');
+    expect(agentsMd).toContain('MCP Server: tooling');
+    expect(agentsMd).toContain('use the tooling server for builds');
+
+    // The full mount set: codex surfaces in, default claude surfaces out.
+    const session = { id: 'session-1', agent_group_id: ag.id } as Session;
+    const config: ContainerConfig = { mcpServers: {}, packages: { apt: [], npm: [] }, additionalMounts: [], skills: [] };
+    const mounts = buildMounts(ag, session, config, 'codex', contribution);
+    const containerPaths = mounts.map((m) => m.containerPath);
+    expect(containerPaths).toContain('/home/node/.codex');
+    expect(containerPaths.some((p) => p.endsWith('AGENTS.md'))).toBe(true);
+    expect(containerPaths).not.toContain('/home/node/.claude');
+  });
+});

src/providers/codex-registration.test.ts

@@ -0,0 +1,27 @@
+/**
+ * Integration test for the codex provider's HOST-side reach-in: the self-registration
+ * import in the src/providers/index.ts barrel. Importing the barrel runs codex.ts's
+ * top-level registerProviderContainerConfig('codex', …); without that import line the
+ * host never wires the provider's per-session mounts / env passthrough.
+ *
+ * Behavior, not structural, and BARREL-ONLY: it imports the real barrel (./index.js),
+ * never ./codex.js directly, then asserts the registry actually contains the provider.
+ * Importing the provider module directly (as codex.factory.test.ts does) self-registers
+ * it and would stay GREEN even if the barrel line were deleted — that is a unit test,
+ * not a registration guard. This test goes red if the barrel import is deleted/drifts,
+ * or the barrel fails to evaluate.
+ *
+ * A provider is a MULTI-POINT integration: this guards the HOST barrel; the CONTAINER
+ * barrel is guarded by the sibling bun test; the SDK/CLI dependency + Dockerfile install
+ * are guarded by the build/container legs (see the skill's validate step).
+ */
+import { describe, it, expect } from 'vitest';
+
+import { listProviderContainerConfigNames } from './provider-container-registry.js';
+import './index.js'; // the real host provider barrel — triggers each provider's self-registration
+
+describe('codex provider host registration', () => {
+  it('registers codex host container-config via the barrel', () => {
+    expect(listProviderContainerConfigNames()).toContain('codex');
+  });
+});

src/providers/codex.ts

@@ -0,0 +1,108 @@
+/**
+ * Host-side container config for the `codex` provider.
+ *
+ * Registers with `providesAgentSurfaces` — codex owns its agent-facing
+ * surfaces, so core skips the default (Claude) compose/mounts and this
+ * contribution supplies them instead:
+ *
+ *   - AGENTS.md — codex's project doc, composed fresh every spawn
+ *     (see ./codex-agents-md.ts), mounted RO over the RW group dir.
+ *   - .agents/skills — codex-native skill links synced to the group's
+ *     container.json selection, mounted RO.
+ *   - ~/.codex — a per-GROUP private state dir (`.codex-shared`), persistent
+ *     across sessions so thread metadata and config.toml survive respawns.
+ *
+ * Credentials: NONE here — v2's invariant is that containers never receive
+ * raw API keys; OneCLI is the sole credential path. The OpenAI key (or
+ * ChatGPT token) lives in the OneCLI vault with an api.openai.com /
+ * chatgpt.com host pattern; codex's traffic already rides the gateway proxy
+ * (every spawn applies it — see container-runner.ts), which injects the real
+ * credential in flight. The container only ever sees the `onecli-managed`
+ * placeholder. Model/effort come from container_config (`ncl groups config
+ * update --model/--effort`), not env.
+ *
+ * Memory and exchange archiving are NOT handled here either — the
+ * container-side provider declares `usesMemoryScaffold` (the runner
+ * scaffolds the memory tree) and implements `onExchangeComplete` (the
+ * provider's own exchange-archive.ts persists each exchange).
+ */
+import fs from 'fs';
+import path from 'path';
+
+import { DATA_DIR } from '../config.js';
+import { getAgentGroup } from '../db/agent-groups.js';
+import { composeGroupAgentsMd } from './codex-agents-md.js';
+import { registerProviderContainerConfig } from './provider-container-registry.js';
+
+registerProviderContainerConfig(
+  'codex',
+  (ctx) => {
+    // Per-group codex state (config.toml, thread metadata).
+    const codexDir = path.join(DATA_DIR, 'v2-sessions', ctx.agentGroupId, '.codex-shared');
+    fs.mkdirSync(codexDir, { recursive: true });
+    // OneCLI bind-mounts its auth stub at ~/.codex/auth.json, nested inside
+    // this dir mount — Docker on macOS can't create a missing mountpoint file
+    // inside a virtiofs bind mount (runc: "mountpoint is outside of rootfs",
+    // exit 125), so it must exist before first spawn. Re-created here per
+    // spawn because a group reset that wipes .codex-shared re-triggers it.
+    // The 'a' flag creates the file if missing, never truncates an existing one.
+    fs.closeSync(fs.openSync(path.join(codexDir, 'auth.json'), 'a'));
+
+    // Compose this group's AGENTS.md and sync codex-native skill links.
+    const group = getAgentGroup(ctx.agentGroupId);
+    if (group) composeGroupAgentsMd(group, ctx.groupDir);
+    syncCodexSkillLinks(ctx.groupDir, ctx.selectedSkills);
+
+    // No credential env here — OneCLI's container-config drives auth end to
+    // end: the gateway serves a sentinel auth.json stub into ~/.codex for
+    // BOTH auth modes (ChatGPT subscription and API key) and swaps the real
+    // credential on the wire. Note the runner's CODEX_ENV_ALLOWLIST
+    // deliberately strips OPENAI_API_KEY from the codex process env — auth
+    // never rides env vars, only the stub. Duplicating any of it here would
+    // be a second source of truth.
+    const mounts = [{ hostPath: codexDir, containerPath: '/home/node/.codex', readonly: false }];
+    const composedAgentsMd = path.join(ctx.groupDir, 'AGENTS.md');
+    if (fs.existsSync(composedAgentsMd)) {
+      // RO over the RW group dir — regenerated every spawn, agent edits would
+      // be clobbered anyway. Memory behavior is edited via memory/system/.
+      mounts.push({ hostPath: composedAgentsMd, containerPath: '/workspace/agent/AGENTS.md', readonly: true });
+    }
+    const agentsDir = path.join(ctx.groupDir, '.agents');
+    if (fs.existsSync(agentsDir)) {
+      mounts.push({ hostPath: agentsDir, containerPath: '/workspace/agent/.agents', readonly: true });
+    }
+
+    return { mounts };
+  },
+  { providesAgentSurfaces: true },
+);
+
+/**
+ * Sync `.agents/skills/<name>` symlinks to the selected skill set. Targets are
+ * container paths (`/app/skills/<name>`) — dangling on the host, valid inside.
+ */
+function syncCodexSkillLinks(groupDir: string, selectedSkills: string[]): void {
+  const skillsDir = path.join(groupDir, '.agents', 'skills');
+  fs.mkdirSync(skillsDir, { recursive: true });
+
+  const desired = new Set(selectedSkills);
+  for (const entry of fs.readdirSync(skillsDir)) {
+    const entryPath = path.join(skillsDir, entry);
+    let isSymlink = false;
+    try {
+      isSymlink = fs.lstatSync(entryPath).isSymbolicLink();
+    } catch {
+      continue;
+    }
+    if (isSymlink && !desired.has(entry)) fs.unlinkSync(entryPath);
+  }
+
+  for (const skill of selectedSkills) {
+    const linkPath = path.join(skillsDir, skill);
+    try {
+      fs.lstatSync(linkPath);
+    } catch {
+      fs.symlinkSync(`/app/skills/${skill}`, linkPath);
+    }
+  }
+}

src/providers/index.ts

@@ -4,3 +4,6 @@
 // needs (claude, mock) don't appear here.
 //
 // Skills add a new provider by appending one import line below.
+
+import './codex.js';
+import './opencode.js';

src/providers/opencode-registration.test.ts

@@ -0,0 +1,27 @@
+/**
+ * Integration test for the opencode provider's HOST-side reach-in: the self-registration
+ * import in the src/providers/index.ts barrel. Importing the barrel runs opencode.ts's
+ * top-level registerProviderContainerConfig('opencode', …); without that import line the
+ * host never wires the provider's per-session mounts / env passthrough.
+ *
+ * Behavior, not structural, and BARREL-ONLY: it imports the real barrel (./index.js),
+ * never ./opencode.js directly, then asserts the registry actually contains the provider.
+ * Importing the provider module directly (as opencode.factory.test.ts does) self-registers
+ * it and would stay GREEN even if the barrel line were deleted — that is a unit test,
+ * not a registration guard. This test goes red if the barrel import is deleted/drifts,
+ * or the barrel fails to evaluate.
+ *
+ * A provider is a MULTI-POINT integration: this guards the HOST barrel; the CONTAINER
+ * barrel is guarded by the sibling bun test; the SDK/CLI dependency + Dockerfile install
+ * are guarded by the build/container legs (see the skill's validate step).
+ */
+import { describe, it, expect } from 'vitest';
+
+import { listProviderContainerConfigNames } from './provider-container-registry.js';
+import './index.js'; // the real host provider barrel — triggers each provider's self-registration
+
+describe('opencode provider host registration', () => {
+  it('registers opencode host container-config via the barrel', () => {
+    expect(listProviderContainerConfigNames()).toContain('opencode');
+  });
+});

src/providers/opencode.ts

@@ -0,0 +1,49 @@
+/**
+ * Host-side container config for the `opencode` provider.
+ *
+ * OpenCode's `opencode serve` process stores state under XDG_DATA_HOME, which
+ * we pin to a per-session host directory mounted at /opencode-xdg. The
+ * OPENCODE_* env vars tell the CLI which provider/model to use at runtime
+ * (read on the host, injected into the container). NO_PROXY / no_proxy are
+ * merged with host values so the in-container OpenCode client can talk to
+ * 127.0.0.1 even when HTTPS_PROXY is set by OneCLI.
+ */
+import fs from 'fs';
+import path from 'path';
+
+import { registerProviderContainerConfig } from './provider-container-registry.js';
+
+function mergeNoProxy(current: string | undefined, additions: string): string {
+  if (!current?.trim()) return additions;
+  const parts = new Set(
+    current
+      .split(/[\s,]+/)
+      .map((s) => s.trim())
+      .filter(Boolean),
+  );
+  for (const addition of additions.split(',')) {
+    const trimmed = addition.trim();
+    if (trimmed) parts.add(trimmed);
+  }
+  return [...parts].join(',');
+}
+
+registerProviderContainerConfig('opencode', (ctx) => {
+  const opencodeDir = path.join(ctx.sessionDir, 'opencode-xdg');
+  fs.mkdirSync(opencodeDir, { recursive: true });
+
+  const env: Record<string, string> = {
+    XDG_DATA_HOME: '/opencode-xdg',
+    NO_PROXY: mergeNoProxy(ctx.hostEnv.NO_PROXY, '127.0.0.1,localhost'),
+    no_proxy: mergeNoProxy(ctx.hostEnv.no_proxy, '127.0.0.1,localhost'),
+  };
+  for (const key of ['OPENCODE_PROVIDER', 'OPENCODE_MODEL', 'OPENCODE_SMALL_MODEL'] as const) {
+    const value = ctx.hostEnv[key];
+    if (value) env[key] = value;
+  }
+
+  return {
+    mounts: [{ hostPath: opencodeDir, containerPath: '/opencode-xdg', readonly: false }],
+    env,
+  };
+});

src/router.ts

@@ -170,7 +170,7 @@ export async function routeInbound(event: InboundEvent): Promise<void> {
       channel_type: event.channelType,
       platform_id: event.platformId,
       name: null,
-      is_group: 0,
+      is_group: event.message.isGroup ? 1 : 0,
       unknown_sender_policy: 'request_approval',
       denied_at: null,
       created_at: new Date().toISOString(),
@@ -289,7 +289,14 @@ export async function routeInbound(event: InboundEvent): Promise<void> {
           log.warn('adapter.subscribe failed', { channelType: event.channelType, threadId: event.threadId, err });
         });
       }
-    } else if (agent.ignored_message_policy === 'accumulate') {
+    } else if (agent.ignored_message_policy === 'accumulate' && !(engages && (!accessOk || !scopeOk))) {
+      // Accumulate stores the message as silent context. We allow it when
+      // engagement simply didn't fire, but NOT when engagement fired and
+      // the access/scope gate refused — those refusals are security
+      // decisions about an untrusted sender, and silently storing their
+      // message (which also stages their attachments to disk via
+      // writeSessionMessage → extractAttachmentFiles) is exactly what the
+      // gate is meant to prevent.
       await deliverToAgent(agent, agentGroup, mg, event, userId, adapter?.supportsThreads === true, false);
       accumulatedCount++;
     } else {

src/session-manager.ts

@@ -14,6 +14,7 @@ import type Database from 'better-sqlite3';
 import fs from 'fs';
 import path from 'path';
 
+import { isSafeAttachmentName } from './attachment-safety.js';
 import type { OutboundFile } from './channels/adapter.js';
 import { DATA_DIR } from './config.js';
 import { getMessagingGroup } from './db/messaging-groups.js';
@@ -252,11 +253,26 @@ function extractAttachmentFiles(
   let changed = false;
   for (const att of attachments) {
     if (typeof att.data === 'string') {
+      // The name field is attacker-controlled: chat platforms with E2E
+      // attachment encryption (WhatsApp, Matrix) cannot sanitize filename
+      // server-side, and other adapters pass att.name through raw. Without
+      // this guard, `path.join(inboxDir, '../../...')` writes anywhere the
+      // host process has fs permission — see Signal Desktop's Nov 2025
+      // attachment-fileName advisory for the same archetype.
+      const rawName = (att.name as string | undefined) ?? `attachment-${Date.now()}`;
+      const filename = isSafeAttachmentName(rawName) ? rawName : `attachment-${Date.now()}`;
+      if (filename !== rawName) {
+        log.warn('Refused unsafe attachment filename — would escape inbox', {
+          messageId,
+          rawName,
+          replacement: filename,
+        });
+      }
       const inboxDir = path.join(sessionDir(agentGroupId, sessionId), 'inbox', messageId);
       fs.mkdirSync(inboxDir, { recursive: true });
-      const filename = (att.name as string) || `attachment-${Date.now()}`;
       const filePath = path.join(inboxDir, filename);
       fs.writeFileSync(filePath, Buffer.from(att.data as string, 'base64'));
+      att.name = filename;
       att.localPath = `inbox/${messageId}/${filename}`;
       delete att.data;
       changed = true;