In an ecosystem of many services, someone has to say when something breaks. Ops Bot is the single point of contact for ecosystem alerts and ops commands, through Telegram.
It receives events from every service (deploy success/failure, health-check breaches, cron failures), categorizes them and delivers them to the operator. The first self-healing L4 pilot of the ecosystem will land here.
- Status
- Live
- Autonomy (AAL)
- L2 → L4
- Channel
- Telegram, two-way
- Intake
- POST /events
01
One inbox for every alert
Services post to a single /events endpoint; the bot sorts events into info, warning, fatal and self_heal.
02
Two-way control
Commands such as status, ack and mute work from Telegram.
03
Home of the L4 pilot
The first self-healing L4 pilot of the ecosystem will land here.
Capabilities
Working now
- POST /events endpoint for service-side notifications
- Telegram bidirectional commands (status, ack, mute)
- Event categorization (info / warning / fatal / self_heal)
Status not stated
- Per-service rate limits and dedup
- Deploy notifications with rollback links
Current autonomy level
Weakest link
No heartbeat — if the bot itself hangs, no one is alerted; logs are unstructured grep-only.
Roadmap to L4
- L3 lift — Munera pings ops bot every 60s, missed beat → SMS fallback; pino JSON logs → Loki; declared tool scopes.
- L4 lift — auto-restart on fatal; classified retry on Telegram 429/5xx; circuit breaker per BOT_TOKEN; hard cost CB on any LLM call in alert generation; pilot integration with self-healing-agents abstractions.
- Verification — 30-day MTBF without manual intervention.
Next step
Explore the available product and its current limits
LIVE means the project is available now; the weakest link and roadmap above state the current version’s limits.
Review the path to the target level