MCP server

Give your AI agent the full picture of a given company: details, customers, partners, vendors, competitors, people and search. Use your API key. Each tool costs the same as its REST endpoint.

Get an API key

1. Set your API key

Keep the key out of version control.

export INNERFETCH_API_KEY="<your API key>"

2. Add the server

Claude Code

Run in a terminal

claude mcp add --transport http innerfetch https://api.innerfetch.com/mcp \
  --header "X-API-Key: $INNERFETCH_API_KEY"

Cursor

.cursor/mcp.json or ~/.cursor/mcp.json

{
  "mcpServers": {
    "innerfetch": {
      "url": "https://api.innerfetch.com/mcp",
      "headers": {
        "X-API-Key": "${env:INNERFETCH_API_KEY}"
      }
    }
  }
}

VS Code

.vscode/mcp.json

{
  "inputs": [
    {
      "type": "promptString",
      "id": "innerfetch-key",
      "description": "Innerfetch API key",
      "password": true
    }
  ],
  "servers": {
    "innerfetch": {
      "type": "http",
      "url": "https://api.innerfetch.com/mcp",
      "headers": {
        "X-API-Key": "${input:innerfetch-key}"
      }
    }
  }
}

3. Add the agent skill

Agent skill

Tells your agent which tool to use, what each call costs and what each error means.

llms.txt

A short overview of the API for language models.

llms-full.txt

Every operation with an example response, plus the price list.

API reference

The full OpenAPI reference.

The skill changes when the API changes. Its version is in the ETag header.