API & Integrations

    What is the Lemwatch MCP server?

    Lemwatch runs a hosted Model Context Protocol server so AI assistants like Claude and ChatGPT can read your sites and act on incidents.

    Updated 2026-07-10
    2 min read
    mcp
    ai
    integrations

    Quick answer#

    MCP (Model Context Protocol) is an open standard that lets AI assistants call live tools. Lemwatch's hosted MCP server exposes 11 tools, 4 resources, and 5 prompts so your AI can list sites, triage incidents, and open handoffs.

    What it does#

  1. Lets Claude / ChatGPT / Cursor read your monitored sites
  2. Streams active incidents into an AI conversation
  3. Exposes safe write actions (ack, comment, handoff) behind confirm gates
  4. Returns fleet health as a first-class resource, not a screenshot

    5. Where to find it in Lemwatch#

    Settings → MCP tokens to issue a token. Docs and endpoint URL live at /mcp.

    How to connect#

  5. Settings → MCP tokens → Issue token. Copy it once — we only store the hash.
  6. In your AI client, add the MCP server URL and paste the token as the bearer.
  7. Ask the assistant "list my sites" to verify.
  8. See the tool catalog for what to try next.

    9. FAQ#

    Which AI clients are supported?#

    Anything speaking MCP 2025-06-18 over Streamable HTTP — Claude Desktop, Cursor, custom agents.

    Is it secure?#

    Tokens are org-scoped, hashed, and rate-limited per plan. Writes require confirm: true.

    Related#

  9. Manage MCP tokens
  10. MCP tool catalog
  11. MCP rate limits by tier
    12. Was this article helpful?

    Previous

    Google Search Console & GA4 Integration

    Next

    How do I create and revoke MCP tokens?

    Related Articles

    How do I create and revoke MCP tokens?

    MCP tokens are org-scoped bearer secrets. Create them in Settings, copy once, and revoke any time — hashes are stored, plaintext never is.

    MCP tool catalog: everything Claude and ChatGPT can call

    The 11 MCP tools, 4 resources, and 5 prompts Lemwatch exposes to AI assistants — with what each one does and who can call it.

    MCP rate limits and tier gating

    Each plan gets a different MCP request-per-minute budget and tool allow-list. Free/LTD-1 read-only and 1 site; Agency gets everything.

    API Endpoints Reference

    Overview of available API endpoints and their usage.

    Back to Help Center