Goal-driven autonomous AI agent that monitors service health, diagnoses incidents, and takes bounded recovery actions.
- Goal-driven agent loop (observe → decide → act)
- Dynamic tool selection
- Log intelligence (RAG over logs)
- Severity classification
- Automatic restart (bounded action)
- Slack alert routing
- Incident memory
- Confidence scoring
- Escalation logic
- Docker-based safe execution
- FastAPI
- LangChain
- LangGraph
- OpenAI API (LLM)
- Qdrant
- SQLite
- Docker
- Docker Compose
- Slack Webhook
- Build your fastapi app image and configure in
tools/container_manager.py
# Create virtual environment
python -m venv venv
# Activate virtual environment
source venv/bin/activatepip install -r requirements.txtuvicorn agent.main:app --reload --port 9000cloudflared tunnel --url http://localhost:9000START
↓
analyze_log (LLM)
↓
decision_router
↙ ↓ ↘
restart slack alerts
↓ ↓ ↓
store_incident_memory
↓
END