1. Home
  2. Servers
  3. MCP Gateways & Proxies
  4. ContextForge

Server profile

ContextForge MCP Server

ContextForge registers and proxies tools, agents, and APIs behind a unified gateway. It targets teams that need discovery, authentication, observability, and policy controls across MCP, A2A, REST, and gRPC services.

View repositoryInstall & configure

Server facts

ContextForge MCP Server

Tech

Runtime / language
pythonPython
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
4,646
Last commit
Data as of

Tags

  • Agent Routing
  • Api Federation
  • Governance
  • Mcp Gateway
  • Observability

Overview

ContextForge is a registry and proxy for bringing different AI-facing services under shared management. Rather than providing one fixed application toolset, it federates MCP servers, A2A agents, and REST or gRPC APIs. Registered backends can be assembled into virtual servers for clients, while common discovery, authentication, and observability controls sit at the gateway layer. The README also describes routing for OpenAI-compatible and Anthropic agents.

The gateway can adapt REST endpoints into MCP tools and discover gRPC services through reflection. Registries cover tools, resources, and prompts, including prompt templates and versioning. Operational features include retries, rate limits, input validation, concurrency controls, and an administrative interface with log inspection. OpenTelemetry support connects tracing to OTLP backends such as Phoenix, Jaeger, or Zipkin. Redis-backed caching and federation support deployments that extend beyond a single gateway process.

The quick start launches the Python package through uvx, with container and Kubernetes deployment paths also described. Configuration includes administrator credentials, UI settings, and token-related secrets; the sample values need customization. After startup, the example registers a backend, inspects its tools, and creates a virtual server before connecting an MCP client. A separate stdio wrapper is available for clients that need process-based access to the HTTP endpoint. The actual tools visible to a client depend on the services registered and the virtual server assembled for it.

Install & configure

Requirements

  • Python 3.11 or newer
  • uv for this launch method
  • Backend services to register
  • curl and jq only for the README smoke test
Documented launch / install command
uvx --from mcp-contextforge-gateway mcpgateway --host 0.0.0.0 --port 4444

This starts the HTTP gateway rather than a stdio MCP process. Customize the provided environment configuration, register backends, and create a virtual server. Its generated identifier determines the client MCP URL. A separate stdio wrapper is documented.

Config file · macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

Windows: %APPDATA%\Claude\claude_desktop_config.json

Use Settings → Developer → Edit Config when available. Do not assume a Linux configuration path; check the client documentation for your platform. Remote servers use the app’s connector flow when supported, not the local stdio file shown here. Merge this example with existing settings; do not overwrite unrelated servers.

No install command stated in the README — see repo. Check the client documentation for a supported connection method.

Read the Claude Desktop setup guide

Config file · Project: .cursor/mcp.json

User: ~/.cursor/mcp.json

The project path is relative to the workspace root. The tilde denotes your user home directory. Use Cursor’s MCP settings to confirm which configuration and scope your installed version is reading. Merge this example with existing settings; do not overwrite unrelated servers.

No install command stated in the README — see repo. Check the client documentation for a supported connection method.

Read the Cursor setup guide

Config file · Workspace: .vscode/mcp.json

For user scope, use the “MCP: Open User Configuration” command if available. Do not guess a user configuration path across operating systems, profiles, containers, or remote workspaces; check the client documentation. Merge this example with existing settings; do not overwrite unrelated servers.

No install command stated in the README — see repo. Check the client documentation for a supported connection method.

Read the VS Code setup guide

Config file · macOS/Linux: ~/.codeium/windsurf/mcp_config.json

Windows: %USERPROFILE%\.codeium\windsurf\mcp_config.json

Open the MCP configuration through Windsurf’s settings when possible to confirm the active file. Paths can differ in managed or future releases; check the client documentation if this location is not present. Merge this example with existing settings; do not overwrite unrelated servers.

No install command stated in the README — see repo. Check the client documentation for a supported connection method.

Read the Windsurf setup guide

Config file · Project scope: .mcp.json at the repository root

The project file uses mcpServers. For user scope, use claude mcp add --scope user rather than assuming .mcp.json in your home directory is the user configuration. Check the client documentation for managed settings and storage details. Merge this example with existing settings; do not overwrite unrelated servers.

No install command stated in the README — see repo. Check the client documentation for a supported connection method.

Read the Claude Code setup guide

Examples are translated from the published configuration. Supply your own credentials and verify requirements in the upstream README. Remote connectors are shown only where a supported configuration format is known.

Tools

No individual tool names were confirmed in the source material. Check the repository for its current tool list.

Names, permissions, and defaults can change. Check the current upstream documentation before granting access.

Use cases

  • Aggregate tools from several MCP servers into a virtual endpoint.
  • Expose an existing REST or gRPC service to MCP clients.
  • Centralize authentication and tracing for assistant-facing services.
  • Manage registered tools, prompts, resources, and agents through an admin interface.

Strengths

  • Federates multiple service and agent protocols.
  • Combines discovery, governance controls, and observability.
  • Supports plugin extensions and deployment beyond a single process.

Limitations

  • Requires backend registration and virtual-server configuration after startup.
  • The available tool catalog depends on connected services.
  • Quick-start credentials and signing secrets require customization.

Frequently asked questions

The README describes registering and federating backend services. Client tool availability follows the configured services and virtual servers.

Yes. REST adaptation and reflection-based gRPC-to-MCP translation are documented.

The README provides a Python stdio wrapper that connects to a configured gateway server URL with authentication.

It uses OpenTelemetry with OTLP-compatible backends, including Phoenix, Jaeger, and Zipkin.

Independent listing, not an affiliation or security review. Grant the least access needed and verify the upstream instructions before running code.

Related MCP servers

ghidraMCP

Security & Reverse Engineering

ghidraMCP connects AI assistants to core Ghidra analysis features through a plugin and Python bridge. It supports binary decompilation, renaming methods and data, and browsing methods, classes, imports, and exports.

javastdio11,310 GitHub stars

Supergateway

MCP Gateways & Proxies

Supergateway bridges MCP transports so stdio servers can reach network clients and remote services can appear locally over stdio. It supports SSE, WebSockets, and Streamable HTTP modes through command-line configuration.

nodestdio2,904 GitHub stars

Unla MCP Gateway

MCP Gateways & Proxies

Unla is a Go gateway that exposes existing APIs and MCP servers through MCP endpoints using configuration. It includes a management interface, recoverable sessions, multi-tenant support, and deployment options for containers or Kubernetes.

nodestdio2,241 GitHub stars

mcp-proxy

MCP Gateways & Proxies

mcp-proxy bridges MCP transport mismatches between clients and servers. It can connect a stdio client to a remote SSE or Streamable HTTP server, or expose local stdio servers through network endpoints.

pythonstdio2,791 GitHub stars

mcpo

MCP Gateways & Proxies

mcpo makes MCP tools available through an OpenAPI-compatible HTTP server. It is for agents and applications that consume OpenAPI endpoints rather than communicating with MCP servers directly.

pythonstdio4,403 GitHub stars

MCPHub

MCP Gateways & Proxies

Organize multiple MCP servers behind shared HTTP or SSE endpoints. MCPHub adds a management dashboard, group-level visibility controls, semantic routing, authentication, and a terminal interface.

nodestdio2,539 GitHub stars

MCP Proxy Server

MCP Gateways & Proxies

MCP Proxy Server combines multiple MCP backends behind one HTTP entrypoint. It aggregates tools, prompts, and resources while connecting to stdio, SSE, or streamable HTTP servers.

gostdio737 GitHub stars

MCP Hub

MCP Gateways & Proxies

MCP Hub coordinates multiple MCP servers behind one client endpoint. It provides a REST management API and web UI, with server lifecycle controls, authentication handling, and live capability updates.

nodestdio521 GitHub stars