generated: '2026-07-19'
method: searched
status: published
source: https://docs.galileo.ai/getting-started/mcp/setup-galileo-mcp
server:
  name: galileo-agent-evals
  transport: http
  url: https://api.galileo.ai/mcp/http/mcp
  auth:
    type: apiKey
    header: Galileo-API-Key
    note: Generate an API key at https://app.galileo.ai/settings/api-keys. Self-hosted deployments use
      a custom base URL.
tools:
- name: create_datasets
  description: Generate synthetic datasets or upload custom data to test and evaluate AI applications
    (general, prompt-injection, off-topic, and toxic-content scenarios).
- name: check_dataset_status
  description: Track dataset generation progress and preview content.
- name: create_prompt_templates
  description: Build reusable prompt templates with model configuration, temperature, and other parameters.
- name: setup_experiments
  description: Configure and run evaluation experiments.
- name: get_signals
  description: Analyze log streams for issues and improvement opportunities.
- name: integrate_openai
  description: Add Galileo observability to OpenAI applications.
- name: integrate_langchain
  description: Enable Galileo tracing for LangChain applications.
- name: search_documentation
  description: Query Galileo documentation for guides and examples.
deployment:
  mode: remote
  endpoint: https://api.galileo.ai/mcp/http/mcp
  verified: probed
  probe: live
  tools: 9
  checked: '2026-08-12'
  source: catalog MCP census