generated: '2026-09-14'
method: probed
source: https://docs.aiola.ai/_mcp/server
name: aiOla Documentation MCP Server
status: live
deployment:
  mode: remote
  endpoint: https://docs.aiola.ai/_mcp/server
  auth: none
  verified: probed
advertised_in:
- https://docs.aiola.ai/llms.txt
- 'every documentation page header: "For AI client integration (Claude Code, Cursor,
  etc.), connect to the MCP server at https://docs.aiola.ai/_mcp/server"'
probe:
  fetched: '2026-09-14'
  initialize:
    http_status: 200
    content_type: text/event-stream
    protocol_version: '2025-11-25'
    server_info:
      name: fern-docs-mcp-server
      version: 1.0.0
    capabilities:
      tools:
        listChanged: true
  tools_list:
    http_status: 200
    tool_count: 1
    saved: mcp/aiola-mcp-tools-list.json
tools:
- name: searchDocs
  description: Search the documentation at https://docs.aiola.ai. Returns relevant doc
    passages with source URLs.
  input_schema:
    query: string
    topK: number
scope:
  covers: documentation search only
  does_not_cover:
  - speech-to-text file transcription
  - speech-to-text streaming
  - text-to-speech synthesis
  - text-to-speech streaming
  - API key to access-token exchange
note: >-
  This is a real, live, unauthenticated remote MCP server on aiOla's own documentation
  host, and it is worth having — an agent can read the docs through it. It is NOT a
  product API server: the single tool searches documentation text. It is generated by
  the Fern documentation platform rather than authored by aiOla, and no aiOla tool can
  transcribe audio, synthesize speech, or mint a token. An agent that needs the Speech
  API still has to call the REST/Socket.IO surface with an API key. No tool crosswalk
  is emitted, because there is no published OpenAPI to bind these tools to and the one
  tool has no backing product operation.