galthran-wq / agent-monitoring-service / 22104509413

88%

Build Type

push

github

Committed by web-flow

Commit Message

feat: implement agent monitoring service (#1)

* feat: implement agent monitoring service

Adds automated infrastructure monitoring that periodically pulls logs from
Loki and metrics from Prometheus, analyzes them via a cheap LLM (Gemini Flash
via OpenRouter), and pushes concise HTML summaries to Telegram.

Components:
- Sources: LokiSource (error/warning log queries), PrometheusSource (up, rates, latency)
- Analyzer: LLM-based analysis with token budget management and statistical fallback
- Exporter: Telegram with edit-previous-message pattern to avoid spam
- Monitor: Async orchestration loop with configurable interval
- API: GET /report endpoint for last generated report

All settings via AGENT_MONITORING_ prefixed env vars. 29 tests with full coverage
of sources, analyzer, exporter, monitor, and API endpoint.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: resolve mypy strict mode errors

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* docs: update README for agent monitoring service

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* Fix Telegram export and monitor settings

* fix: sort imports in telegram exporter

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Cursor Agent <cursoragent@cursor.com>

Run Details

317 of 358 new or added lines in 15 files covered. (88.55%)

1 existing line in 1 file now uncovered.

404 of 454 relevant lines covered (88.99%)

0.89 hits per line