#77 feat: Slack notifications for fleet state changes (verify, blocked, offline)

[Anonymous]

Originally created by: kumaakh

A standalone watcher process that reads statusline-state.json periodically and POSTs to a Slack webhook when member state changes.

Target events: member hits verify checkpoint, member becomes blocked, member goes offline unexpectedly.

Design notes:

• Watcher reads statusline-state.json (already written by the statusline service)
• Configurable webhook URL per fleet
• Subsumes the inter-session attention mechanism (#14) for teams using Slack
• Should be opt-in, not required

Backlog item [#18] from docs/MCP-BACKLOG.md. Future.

[Anonymous]

Originally posted by: kumaakh

Technical direction: This depends on the status line infrastructure proposed in [#149]. Recommended sequencing:

1. Implement publish_status tool (#149) first — the watcher's state source is statusline-state.json which that tool maintains.
2. The Slack watcher is then a simple standalone process (scripts/fleet-slack-watcher.mjs or a new service in src/services/slack-notifier.ts) that:
3. Reads statusline-state.json on a configurable poll interval (e.g. 30s)
4. Diffs member states against previous read
5. POSTs to a configured Slack webhook URL on state change to erify, locked, or member going offline
6. Webhook URL stored in ~/.apra-fleet/config.json (or an env var) — never in the fleet registry.
7. Opt-in: the notifier only runs if webhook URL is configured; no-op otherwise.

Consider whether this should be a long-running child process started by the MCP server, or a separate CLI command (pra-fleet slack-watch) the user runs independently. The latter is simpler and avoids the MCP server managing additional processes.