Skip to main content
Gemini CLI connects to Equip’s MCP server at https://equip.co/api/v3/mcp/ through its settings file. The server is read-only, so Gemini CLI can look things up but never change them.

​
Add the Server

You need your organization’s API key in your shell as EQUIP_MCP first; see Connect Your AI Assistant to Equip.
1

Edit settings.json

Add an equip entry under mcpServers in ~/.gemini/settings.json (or .gemini/settings.json inside one project):
Gemini CLI expands $EQUIP_MCP from your environment when it loads the file, so the key itself never sits in it. Use httpUrl, not url: url is for the older SSE transport, which Equip does not serve.
2

Check the connection

Start Gemini CLI and run /mcp. equip is listed with its tools.
3

Ask a question

Try: “Which of my job openings have the most candidates waiting in screening?”
Edit the file rather than running gemini mcp add --header: the command stores the header value literally, so your key would be written into the settings file.
  • 401 Invalid API key: the key was mistyped or rotated, or EQUIP_MCP is not exported in the shell you started Gemini CLI from. An unset variable expands to an empty string, so run echo $EQUIP_MCP there.
  • 403 “isn’t included in your plan”: your plan does not include the MCP server. See Compare Plans, or contact support to have it switched on for your organization.
  • The server shows but has no tools: quit and restart Gemini CLI after editing the file.