Apollo.io MCP Server

What is Apollo.io MCP Server?

An MCP server that exposes the Apollo.io API functionalities as tools, allowing interaction with the Apollo.io API via the Model Context Protocol (MCP). It is intended for developers who want to integrate Apollo.io data enrichment and search capabilities into MCP-compatible clients (e.g., Claude for Desktop).

How to use Apollo.io MCP Server?

Set the APOLLO_IO_API_KEY environment variable (or create a .env file with it), install dependencies with uv sync, then run uv run mcp run server.py. For use with Claude for Desktop, add a configuration entry to claude_desktop_config.json using stdio transport and the uv run mcp run command pointing to server.py.

Key features of Apollo.io MCP Server

• People enrichment: enrich data for one person.
• Organization enrichment: enrich data for one company.
• People search: find people via Apollo.io.
• Organization search: find organizations via Apollo.io.
• Organization job postings: find job postings for a specific organization.

Use cases of Apollo.io MCP Server

• Enrich contact records with detailed person or company data.
• Search for potential leads or target accounts by people or organization criteria.
• Retrieve current job postings from a specific company for recruitment or market analysis.

FAQ from Apollo.io MCP Server

What Apollo.io API endpoints are exposed as tools?

Five tools are available: people_enrichment, organization_enrichment, people_search, organization_search, and organization_job_postings.

What runtime dependencies are required?

The server requires Python, uv (for sync and running), and the packages listed (implicitly from uv sync). An Apollo.io API key is mandatory.

How does authentication work?

Authentication is via an Apollo.io API key set as the environment variable APOLLO_IO_API_KEY or in a .env file in the project root.

Can this server be used with Claude for Desktop?

Yes. It uses the stdio transport. A sample configuration is provided in the README for claude_desktop_config.json.

Are there any known limitations or data residency considerations?

The README does not mention any limitations or where data is processed. Data is sent to and from the Apollo.io API based on the user's queries.