Overview
What is Public Apis Mcp?
Public Apis Mcp is an MCP server that provides a catalog of free public APIs with semantic search. It uses the public-apis GitHub repo as its data source and is designed for developers who want to discover and query public APIs via MCP tools.
How to use Public Apis Mcp?
Install using uvx public-apis-mcp and add the server configuration to an MCP client (e.g., Claude Desktop). Run the server with uvx public-apis-mcp run; optionally build the embedding index first with uvx public-apis-mcp build-index. Two tools are provided: search_public_apis for semantic search and get_public_api_details to retrieve full details by ID.
Key features of Public Apis Mcp
- Embedding-based semantic search over API names and descriptions
- Retrieve full API details by unique ID
- Resources:
public-apis://apis,public-apis://api/{id} - Auto-builds embedding index on first search
- Testable with MCP Inspector
Use cases of Public Apis Mcp
- Discovering public APIs for a project using natural language queries
- Getting detailed information about a specific API from the catalog
- Integrating API discovery into AI assistants via MCP tools
FAQ from Public Apis Mcp
How do I install Public Apis Mcp?
Use uvx public-apis-mcp or add the server config to your MCP client using uv.
What data does Public Apis Mcp use?
It uses the public-apis GitHub repo, stored locally in src/public_apis_mcp/datastore/free_apis.json.
Does Public Apis Mcp require any external dependencies?
Python 3.10+ and uv are required for development; runtime uses uvx.
How do I test Public Apis Mcp?
Set FREE_APIS_MCP_TEST_MODE=1 and run uv run pytest -q. You can also use MCP Inspector via npx @modelcontextprotocol/inspector uv run public-apis-mcp.
What transport does Public Apis Mcp use?
It runs over STDIO.
