{"id":"mcp-link","name":"MCP Link","homepage":"https://github.com/automation-ai-labs/mcp-link","repo_url":"https://github.com/automation-ai-labs/mcp-link","category":"developer-tools","subcategories":["api-integration","openapi","automation"],"tags":["openapi","rest-api","api-bridge","go","automation","github","slack","stripe","notion"],"what_it_does":"Converts any OpenAPI V3 REST API specification into a working MCP server automatically, enabling AI agents to call any REST API through MCP without writing custom integration code.","use_cases":["Instantly make any OpenAPI V3 API (GitHub, Slack, Stripe, Notion, Figma, YouTube) accessible to Claude without custom code","Build AI agent workflows that call multiple REST APIs through a unified MCP interface","Selectively expose a subset of API endpoints to an agent using path filtering expressions","Rapidly prototype AI integrations against internal APIs that have OpenAPI specs","Replace manual curl/fetch calls in agent workflows with structured MCP tool calls"],"not_for":["APIs without an OpenAPI V3 specification (OpenAPI V2/Swagger requires conversion first)","APIs requiring OAuth interactive flows (not currently supported)","APIs with complex MIME types or multipart file uploads (limited MIME support noted)"],"best_when":"You have a REST API with an OpenAPI V3 spec and want to give an AI agent access to it immediately without building a custom MCP server. Especially powerful for well-documented public APIs.","avoid_when":"Your API uses OAuth 2.0 authorization code flow, requires interactive auth, or relies on resource-based interactions not expressible in OpenAPI. Also avoid for APIs without a spec.","alternatives":["mcp-graphql","custom-mcp-server","langchain-api-tools"],"af_score":62.0,"security_score":60.0,"reliability_score":null,"package_type":"mcp_server","discovery_source":["github"],"priority":"low","status":"evaluated","version_evaluated":"latest","last_evaluated":"2026-03-01T09:50:05.872421+00:00","performance":{"latency_p50_ms":null,"latency_p99_ms":null,"uptime_sla_percent":null,"rate_limits":null,"data_source":"llm_estimated","measured_on":null}}