maps-geo MCP Server
Production-grade multi-agent AI DevOps system for SRE incident response, where a central orchestrator coordinates 12 specialist agents to classify incidents, traverse service dependency graphs, analyze repositories, Terraform, and telemetry, determine root cause, and stream real-time remediation plans to operators via SSE.
Discovered via github-topic:mcp and last synced 3mo ago.
Install instructions not detected yet
Check the source repository for the latest setup steps.
Technology
Prompt injection detection, PII scrubbing (Presidio + regex fallback)
Server-Sent Events (SSE) via sse-starlette
Stream closed — session_id echoed for confirmation
Every hour
Yes\*
No
No
Push / PR
PostgreSQL 16 + SQLAlchemy 2 async + Alembic migrations
GitHub MCP server, Terraform MCP server, custom Ops Inspector server
Root Cause Finder result — causal chain, timeline, confidence
Schedule
Yes
No
No
Trigger
Python 3.11, FastAPI, LangGraph, CrewAI, LangChain
Redis 7 + Celery (periodic graph maintenance tasks)
Next.js 15, TypeScript, Tailwind CSS, Framer Motion, anime.js, Jest
Emitted first — contains the `session_id` for this analysis turn
Daily 02:00
No
Push / PR
JWT (access + refresh tokens), bcrypt password hashing, python-jose
Stream-level error (guardrail violation, unexpected exception)
Yes\*
No
Docker Compose — dev and prod overlays
Every 15 min
No
Push / PR
OpenAI (default) · Anthropic · Google — runtime-switchable, no rebuild
Graph Analyzer result — blast-radius, dependencies, runbooks
Required
No
Push to main
Agent lifecycle update (start, complete, error, skipped)
MCP Server
No
Push to main
Final combined output — natural response, structured data, citations
Yes\*
No