Sandbox
@automation-ai-labs/mcp-link

OpenAPI to MCP server adapter

MCP Link converts an OpenAPI V3 spec into an MCP server that agents can call. It reads the API schema, maps endpoints into MCP tools, and serves them over SSE with optional auth and path filters.

623 stars67 forksGoUpdated 1y ago
Who it's for

Builders who want to expose REST APIs to Claude Code, Codex, Cursor, or other MCP clients.

What it delivers

You can turn an existing API into an agent-callable MCP server without hand-writing each interface.

What it does

Automatic OpenAPI conversion

Generates MCP server behavior from an OpenAPI schema instead of manual tool definitions.

SSE MCP transport

Serves the converted API over an SSE endpoint that MCP clients can connect to.

Auth header support

Accepts an authentication header format in the connection URL for protected APIs.

Path filtering

Lets you include or exclude endpoints with wildcard filters such as `+/**:GET;-/internal/**`.

Sample API configs

Includes example YAML files for APIs like Brave, DuckDuckGo, Firecrawl, Notion, Slack, and YouTube.

How to get it

  1. 1Run
    # Clone repository
    git clone https://github.com/automation-ai-labs/mcp-link.git
    cd mcp-openapi-to-mcp-adapter
    
    # Install dependencies
    go mod download
  2. 2Run
    # Specify port
    go run main.go serve --port 8080 --host 0.0.0.0

README

MCP Link - Convert Any OpenAPI V3 API to MCP Server

Join our Discord

🧩 Architecture

MCP Link

🤔 Why MCP Link?

There is a notable gap in the current AI Agent ecosystem:

  • Most MCP Servers are simple wrappers around Web APIs
  • Functionality interfaces may not be complete, depending on developer implementation
  • Manual creation of MCP interfaces is time-consuming and error-prone
  • Lack of standardized conversion processes

MCP Link solves these issues through automation and standardization, allowing any API to easily join the AI-driven application ecosystem.

🌟 Key Features

  • Automatic Conversion: Generate complete MCP Servers based on OpenAPI Schema
  • Seamless Integration: Make existing RESTful APIs immediately compatible with AI Agent calling standards
  • Complete Functionality: Ensure all API endpoints and features are correctly mapped
  • Zero Code Modification: Obtain MCP compatibility without modifying the original API implementation
  • Open Standard: Follow the MCP specification to ensure compatibility with various AI Agent frameworks

🌐 Online Version

Try our hosted version at mcp-link.vercel.app to quickly convert and test your APIs without installation.

🚀 Quick Start

Installation

# Clone repository
git clone https://github.com/automation-ai-labs/mcp-link.git
cd mcp-openapi-to-mcp-adapter

# Install dependencies
go mod download

Running

# Specify port
go run main.go serve --port 8080 --host 0.0.0.0

Parameter Description

  • s= - URL of the OpenAPI specification file
  • u= - Base URL of the target API
  • h= - Authentication header format, in the format of header-name:value-prefix
  • f= - Path filter expressions to include or exclude API endpoints. Syntax:
    • +/path/** - Include all endpoints under /path/
    • -/path/** - Exclude all endpoints under /path/
    • +/users/*:GET - Include only GET endpoints for /users/{id}
    • Multiple filters can be separated by semicolons: +/**:GET;-/internal/**
    • Wildcards: * matches any single path segment, ** matches zero or more segments

Examples

_APIMCP Link URLAuthentication Method
BraveBrave Searchhttps://mcp-link.vercel.app/links/braveAPI Key
DuckDuckGoDuckDuckGohttps://mcp-link.vercel.app/links/duckduckgoNone
FigmaFigmahttps://mcp-link.vercel.app/links/figmaAPI Token
GitHubGitHubhttps://mcp-link.vercel.app/links/githubBearer Token
Home AssistantHome Assistanthttps://mcp-link.vercel.app/links/homeassistantBearer Token
NotionNotionhttps://mcp-link.vercel.app/links/notionBearer Token
SlackSlackhttps://mcp-link.vercel.app/links/slackBearer Token
StripeStripehttps://mcp-link.vercel.app/links/stripeBearer Token
TMDBTMDBhttps://mcp-link.vercel.app/links/tmdbBearer Token
YouTubeYouTubehttps://mcp-link.vercel.app/links/youtubeBearer Token

Usage in AI Agents

{
  "mcpServers": {
    "@service-name": {
      "url": "http://localhost:8080/sse?s=[OpenAPI-Spec-URL]&u=[API-Base-URL]&h=[Auth-Header]:[Value-Prefix]"
    }
  }
}

These URLs allow any API with an OpenAPI specification to be immediately converted into an MCP-compatible interface accessible to AI Agents.

📋 Future Development

  • MCP Protocol OAuthflow: Implement OAuth authentication flow support for MCP Protocol
  • Resources Support: Add capability to handle resource-based API interactions
  • MIME Types: Enhance support for various MIME types in API requests and responses

Files in the repo

Repository payload9 top-level entries
  • assets
  • examples
  • utils
  • .gitignore
  • go.mod
  • go.sum
  • LICENSE
  • main.go
  • README.md

Discussion (0)

Ask about usage, or say what you built with it

Sign in to join the discussion.

No comments yet. Be the first to say what this is good for.

More connectors

High-performance code intelligence MCP server. Indexes codebases into a persistent knowledge graph — average repo in milliseconds. 158 languages, sub-ms queries, 99% fewer tokens. Single static binary, zero dependencies.

43k

Universal provider proxy for OpenAI Codex & Claude Code — use any LLM (Claude, Gemini, Grok, DeepSeek, Ollama…) with Codex CLI, App, SDK, and Claude Code

14k
okf-memory/
okf-agent-memory

Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.

547
tirth8205/
code-review-graph

Local-first code intelligence graph for MCP and CLI. Builds a persistent map of your codebase so AI coding tools read only what matters, with benchmarked context reductions on reviews and large-repo workflows.

31k
2akouwu/
reverify

Stop your AI from making things up — it proposes, deterministic tools decide, every claim checked against ground truth with evidence. Grounded facts and context survive resets. Reverse engineering is the proving ground. MCP server + CLI.

1.1k
t8y2/dbxConnectors

20 MB lightweight cross-platform database client for 90+ databases, including MySQL, PostgreSQL, SQLite, Redis, MongoDB, DuckDB, SQL Server, and Dameng. Built-in AI, MCP Server, CLI, desktop and Docker. | 轻量级跨平台数据库管理工具,支持 MySQL、PostgreSQL、SQLite、Redis、MongoDB、达梦等 90+ 数据库,提供桌面端、Docker、CLI、内置 AI 助手和 MCP Server。

19k