Skip to main content

Connect Other MCP Clients

:::caution Experimental The 32 MCP tools work over streamable-http from any MCP client (Cursor, Codex, custom Agent SDK) — but the SignalPilot skills are Claude Code-specific and don't run outside it. You'll have the tools without skill orchestration. The Claude Code plugin is the supported path; treat the configs below as best-effort. :::

Claude Desktop​

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

claude desktop config
{
  "mcpServers": {
    "signalpilot": {
      "type": "http",
      "url": "http://localhost:3300/mcp"
    }
  }
}

For cloud with bearer auth:

claude desktop cloud config
{
  "mcpServers": {
    "signalpilot": {
      "type": "http",
      "url": "https://gateway.signalpilot.ai/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Cursor​

Add to .cursor/mcp.json in your project root:

cursor config
{
  "mcpServers": {
    "signalpilot": {
      "type": "http",
      "url": "http://localhost:3300/mcp"
    }
  }
}

Codex / OpenAI Agent SDK​

codex / openai config
{
  "mcpServers": {
    "signalpilot": {
      "type": "http",
      "url": "http://localhost:3300/mcp"
    }
  }
}

Custom Agent SDK​

Any client implementing MCP streamable-http transport can connect:

  • Local endpoint: http://localhost:3300/mcp
  • Cloud endpoint: https://gateway.signalpilot.ai/mcp
  • Auth header: Authorization: Bearer YOUR_API_KEY

The gateway returns standard MCP responses. All 32 tools are callable — governance runs on the server side regardless of which client you use.

Bearer token auth​

When API keys are configured, pass them as a bearer token:

bearer token
Authorization: Bearer sk_your_key_here

For local self-hosted deployments without API key enforcement, no auth header is required.

Limitations for non-Claude clients​

  • No skills — dbt-workflow, sql-workflow, and dialect-specific skills are Claude Code-specific. You'll need to craft your own prompt instructions.
  • No verifier agent — the post-build 7-check protocol is a Claude Code plugin agent. It won't run.
  • No auto-skill loading — you must manually structure your prompts to call the right tools in the right order.

For full functionality, use Claude Code.