Monitor application logs for errors and generate JSON incident files with Telegram alerts.
npx skills add https://github.com/fluxforgeai/ARTEMIS --skill watchdog-fluxforgeai
Please help me install this Agent Skill. Skill: watchdog Source: https://github.com/fluxforgeai/ARTEMIS/tree/main/.claude/skills/watchdog Command: npx skills add https://github.com/fluxforgeai/ARTEMIS --skill watchdog-fluxforgeai
Provides an autonomous background monitor that watches application logs for errors after a fix, creates structured incident files, de-duplicates repeated errors, and pushes timely Telegram alerts so teams can receive immediate notifications without keeping sessions open.
Start the watchdog to monitor the backend container for the recent fix by invoking: /watchdog Monitor extraction after retry logic fix - watch for timeout errors
High-intent search queries and answers about installing and using this skill.
To monitor application logs for errors with Telegram alerts, run an autonomous bash daemon that polls logs, tags incidents as system or fix-related, and sends push notifications directly to a configured Telegram chat.
Run a background monitoring process after deploying a fix by starting a bash daemon that polls logs every 60 seconds, de-duplicates repeated errors, and alerts you to regressions even after the session lifecycle ends.
Yes, you can de-duplicate repeated application errors to avoid alert spam by hashing error signatures, which prevents duplicate notifications while still writing structured JSON incident files for each unique error.
The best way to verify a production fix without keeping a terminal session open is using an independent background bash daemon that continuously monitors container logs and sends push notifications for regressions.
Yes, the log watchdog supports containerized backends, allowing you to monitor application logs within containers for errors and receive immediate Telegram notifications for post-fix verification.
You can check incident status without parsing raw logs by reading a generated status JSON file that surfaces current monitoring state, allowing quick CLI or AI-driven checks for recent system or fix-related incidents.
HahyeonJeonConfigure Telegram bot token and chat ID for Claude Code notifications.
Architekt-JutraEvaluates AI agent incident diagnoses for latency degradation events using predefined rubrics and artifact files.
nicolehaugenValidate CI/CD pipelines for semver, token types, retry logic, and draft detection.
chiptoe-svgAdd a local monitoring dashboard for NanoClaw deployments.
