Commit-Relay MCP server
Commit Nov 5, 2026An autonomous multi-agent platform that manages the GitHub repository lifecycle through intelligent task routing and self-healing workers.
Also known as ry-ops/commit-relay
- Stars
- 2
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- —
- Last commit
- Nov 5, 2026
- License
- —
What is Commit-Relay MCP?
Commit-Relay is an autonomous multi-agent platform that manages the GitHub repository lifecycle through intelligent task routing and self-healing workers.
Summary generated from the project's README
About Commit-Relay MCP server
Managing the entire GitHub repository lifecycle through autonomous agents is the primary function of this archived platform. It uses a central Coordinator and a Mixture of Experts router to distribute tasks to specialized master agents, such as Security, Development, and CI/CD masters. These masters then spawn specific workers for implementation, testing, or documentation to handle repository tasks without manual intervention.
Developers can interact with the system via an MCP server that exposes capabilities as tools, or through a Python SDK. The architecture relies on a multi-provider LLM mesh supporting Anthropic Claude, OpenAI, and Ollama, utilizing cost-aware model selection and automatic failover. Because the project is archived, it is intended for architectural reference rather than active deployment due to unmaintained dependencies and known vulnerabilities.
System reliability is maintained through self-healing daemons that perform heartbeat monitoring, zombie worker cleanup, and failure detection. The platform also incorporates a RAG system using FAISS vector stores to provide context from code, documentation, and decision logs.
What you can do with it
- 1Automated repository maintenanceAn AI client can trigger the full lifecycle of a GitHub repository from task routing to deployment.
- 2Intelligent task distributionThe system routes incoming requests to the correct specialized agent using a keyword-based router.
- 3Multi-model LLM orchestrationThe platform manages different AI providers to balance cost and complexity for various tasks.
Written from the project's README. Every use case is checked against a quote from it.
Automated checks
Ran Nov 16, 2026Full report →Not verified yet. A server needs a score of 80, no serious finding, and enough data for us to check. How checks work · What to fix for more points
- Source repository linked and reachablery-ops/commit-relay is public and reachablegithub.com/ry-ops/commit-relay ↗Pass
- Package names this repository as its sourceNot checked
- Build provenance (Sigstore attestation)Not checked
- Open source license (OSI)No licenseFinding
- Commit or release in the last yearLast commit or release 2026-11-05github.com/ry-ops/commit-relay ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNot checked
- No critical advisory in the package or its dependenciesNot checked
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubCommit-Relay
Multi-agent AI system for autonomous GitHub repository management. Archived, and kept for reference.
[!IMPORTANT] Archived. Commit-Relay is no longer developed or run anywhere, and the code is kept as-is for reference. Its dependencies are not maintained and have known vulnerabilities, so read it, borrow ideas from it, but don't deploy it.
What is Commit-Relay?
Commit-Relay was an autonomous multi-agent platform that managed the entire GitHub repository lifecycle — from task routing to code implementation, security scanning, testing, and deployment — with zero manual intervention.
Agents communicated through structured coordination files rather than direct API calls, creating a fully auditable, transparent orchestration system. A central Coordinator routed incoming tasks to specialized master agents using a Mixture of Experts (MoE) router, which then spawned lightweight workers to execute in parallel.
Current state
| Status | Archived: no development, no running deployment |
| Last functional change | March 2026 (animated architecture docs); later commits are dependency and security housekeeping |
| Dependencies | Unmaintained, with known vulnerabilities. Not safe to deploy as-is. |
| What's useful today | The architecture: file-based agent coordination, MoE task routing, self-healing workers, and the multi-provider LLM gateway |
| Size | About 2,300 files: coordination state and agent outputs, 200+ scripts, an Express API server, a Python SDK and an MCP server |
Architecture
Core Components
| Component | Description |
|---|---|
| Master Agents | Five core masters (Coordinator, Development, Security, Inventory, CI/CD), plus Achievement and Aggregator |
| 7 Worker Types | Implementation, Fix, Test, Scan, Security Fix, Documentation, Analysis |
| 8+ Autonomous Daemons | Coordinator, Worker Manager, Process Monitor, Heartbeat, Zombie Cleanup, Worker Restart, Failure Detection, Auto-Fix |
| LLM Mesh Gateway | Multi-provider support (Anthropic, OpenAI, Ollama) with circuit breakers, cost tracking, and automatic failover |
| MoE Router v4.0 | Keyword activation with learned weights and semantic routing. The project's docs describe 350+ activation keywords. |
| RAG System | FAISS vector store with 5 collections (code, docs, decisions, patterns, tasks) using sentence-transformers |
| API Server | Express.js with 100+ REST endpoints, WebSocket streaming, rate limiting and authentication |
| MCP Server | Model Context Protocol interface exposing system capabilities as tools |
| Python SDK | Full client library with task orchestration, analytics, health monitoring, and reporting |
| Observability | Elastic APM, LangSmith tracing, 27 event types, distributed tracing, anomaly detection |
Key Capabilities
Intelligent Task Routing
- Mixture of Experts router with 350+ keywords and continuous learning
- PyTorch neural routing with training pipeline
- Semantic routing via embeddings (94.5% coverage)
- Margin-based confidence with automatic fallback
Self-Healing System
- 12+ automated remediation strategies
- Heartbeat monitoring with 2-minute intervals
- Zombie worker detection and cleanup
- Exponential backoff restart logic
- ML-based failure pattern recognition
LLM Mesh (Multi-Provider Gateway)
- Anthropic Claude (primary), OpenAI, and Ollama support
- Cost-aware model selection (simple tasks -> haiku, complex -> opus)
- Circuit breaker middleware with provider health monitoring
- Automatic failover chains with retries
- Token usage and cost analytics
Enterprise Governance
- PII scanning (emails, phone numbers, SSNs, API keys)
- RBAC with permission inheritance across 7 namespaces
- SOC2, GDPR, HIPAA compliance policy checking
- Data quality monitoring with schema validation
- Complete audit trail via file-based coordination
Observability Stack
- Elastic Cloud APM with custom spans and business metrics
- LangSmith for LLM performance tracking
- 27 event types with real-time streaming
- Distributed tracing with waterfall visualization
- 50+ system metrics with aggregation
RAG-Enhanced Context
- Pluggable vector store (Weaviate, Qdrant, file-based)
- Connectors for GitHub, Confluence, Slack
- Hybrid search (BM25 + semantic with Reciprocal Rank Fusion)
- 5 specialized collections with metadata schemas
Project Structure
commit-relay/
├── agents/ # Agent configs, prompts, logs, worker outputs
├── api-server/ # Express.js API server (100+ endpoints)
├── config/ # System configuration
├── coordination/ # File-based coordination (task queue, worker pool, handoffs)
│ ├── masters/ # Master agent configurations and libraries
│ ├── governance/ # Governance policies and audit logs
│ ├── catalog/ # Data and AI catalog
│ └── observability/ # Event streams and metrics
├── docs/ # 60+ documentation files and the diagrams on this page
├── examples/ # Usage examples
├── lib/ # Shared libraries
│ ├── cache/ # Adaptive LRU cache
│ ├── governance/ # PII scanner, access control, compliance
│ ├── orchestration/ # Workflow engine, SLA monitor, rate limiter
│ └── rag/ # Vector store, embeddings, connectors
├── llm-mesh/ # Multi-provider LLM gateway
├── mcp-server/ # Model Context Protocol server
├── python-sdk/ # Python client library
├── scripts/ # 200+ operational scripts
├── security/ # Security scanning and CVE tracking
└── testing/ # Test suites and test utilities
Tech Stack
| Category | Technologies |
|---|---|
| Runtime | Node.js 18+, Python 3.8+, Bash |
| AI/ML | Anthropic Claude, OpenAI, Ollama, PyTorch, sentence-transformers, FAISS |
| API | Express.js 5, WebSocket, JSON-RPC 2.0 (MCP) |
| Observability | Elastic APM, LangSmith, OpenTelemetry |
| Security | Helmet, express-rate-limit, JWT, PII scanning |
| Data | FAISS, Weaviate, Qdrant, MiniSearch |
| Deployment | Docker, systemd/launchd |
License
Commit-Relay MCP questions
Is Commit-Relay still being developed?
No, the project is archived and no longer under development or running in any deployment.
What LLMs does the system support?
The LLM Mesh Gateway supports Anthropic Claude, OpenAI, and Ollama.