mcp-context-forge

mcp-context-forge

An AI gateway, registry, and proxy that unifies management of MCP, A2A, or REST/gRPC APIs.

MCPDevOpen sourceAI GatewayMCPA2ARESTgRPC
Type
MCP
Transport
http
Open source
Yes
GitHub Stars
★ 3.9k
Source
mcp-github

Overview

ContextForge is an open-source registry and proxy that consolidates any MCP, A2A, or REST/gRPC API into a single unified endpoint, providing centralized governance, discovery, and observability. It optimizes proxying and tool invocation, and supports plugin extensions. With ContextForge, AI clients can access multiple backend services through a single interface, simplifying management and configuration. Ideal for enterprises and developers needing to integrate multiple AI services and APIs.

Capabilities

  • MCP and A2A protocol support
  • REST/gRPC to MCP translation
  • Centralized governance and discovery
  • OpenTelemetry observability
  • Multi-cluster federation and caching

Use cases

Integrating multiple AI servicesUnified management of APIs and proxiesImproving observability in AI systemsSimplifying AI infrastructure configuration

Setup

Requires: Python环境Docker(可选)Redis(可选)
pip install mcp-contextforge-gateway or run with Docker: docker run -p 8080:8080 ghcr.io/ibm/mcp-context-forge

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

How do I get started quickly?

Refer to the 5-minute quick start guide: https://github.com/IBM/mcp-context-forge/issues/2503

What transport protocols are supported?

Supports HTTP, JSON-RPC, WebSocket, SSE, and more

Related skills