future-audiences/wikimedia-enterprise-model-context-protocol

{
  "service_type": "platform",
  "base_url": "https://gitlab.wikimedia.org",
  "auth_method": "openid_connect",
  "auth_config": {
    "authorization_url": "https://gitlab.wikimedia.org/users/auth/openid_connect",
    "token_url": "https://gitlab.wikimedia.org/oauth/token"
  },
  "endpoints": [],
  "pricing_model": {
    "type": "unknown",
    "details": {
      "note": "Part of Wikimedia Enterprise offering; pricing may be subscription-based or based on usage tiers."
    }
  },
  "rate_limits": {},
  "capabilities": [
    "Wikipedia article lookup",
    "MediaWiki API integration",
    "Model Context Protocol (MCP) server implementation",
    "Python-based service",
    "Hosted on GitLab with OpenID Connect authentication",
    "Part of Wikimedia Enterprise future audiences project"
  ],
  "raw_analysis": "This is a GitLab repository hosted at gitlab.wikimedia.org under the 'future-audiences' group, specifically a project named 'wikimedia-enterprise-model-context-protocol'. The description indicates it is a 'Wikipedia Article lookup API' with emojis suggesting a Python (🐍) and cloud (☁️) implementation. The platform appears to be a Model Context Protocol (MCP) server that provides an API for looking up Wikipedia articles, likely aiming to integrate with language models or other AI tools via the MCP standard. The repository is part of the Wikimedia Enterprise initiative, which offers commercial access to Wikimedia project data. The HTML content suggests that the page triggers an OpenID Connect authentication flow automatically, indicating that the GitLab instance is secured via SSO. The project seems to be in an early-stage or experimental phase (under 'future-audiences'), and the actual API endpoints are not exposed publicly; they would be served by the MCP server when deployed. The platform is likely aimed at developers or organizations needing programmatic access to Wikipedia content, possibly for AI-assisted research, summarization, or knowledge retrieval. Maturity is low, given the experimental label, but it leverages Wikimedia's established infrastructure. There are no visible integration details beyond the MCP protocol. The GitLab repository itself does not present a public API; the actual service would run elsewhere and require authentication tokens. Pricing is unclear but may align with Wikimedia Enterprise subscription models. Rate limits would be defined by the service implementation, not visible here. Access to the repository source code might be restricted or require a Wikimedia account."
}

3/3 tests passed

{
  "overall": 48,
  "dimensions": {
    "token_efficiency": 5.0,
    "first_try_success": 4.0,
    "response_parseability": 8.0,
    "error_clarity": 5.0,
    "doc_quality": 6.0,
    "auth_simplicity": 3.0,
    "latency": 8.0,
    "consistency": 5.0
  },
  "pricing_normalized": {
    "type": "unknown",
    "note": "Part of Wikimedia Enterprise; pricing not disclosed publicly. Agents cannot confidently recommend without clear cost structure."
  },
  "issues": [
    "Auth requires OpenID Connect, which is not as simple as magic link or SSO for agents to onboard; user must go through GitLab institutional sign-in.",
    "No llms.txt file returned; only HTML login form for GitLab, so documentation is not readily available for automated parsing.",
    "Platform appears to be an internal/experimental project (part of Wikimedia Enterprise future audiences), not a stable consumer product.",
    "Pricing model unknown and not discoverable from website; could not validate cost transparency.",
    "The robots.txt has moderation and may restrict some crawling, adding friction for agent discovery.",
    "LLM consumption requires MCP integration, which is not yet a standard; increases setup complexity for agents."
  ],
  "recommendations": [
    "Publish a dedicated llms.txt with clear API and MCP documentation for automated onboarding.",
    "Provide SSO or magic-link authentication to reduce friction and allow agents to onboard independently.",
    "Clarify pricing model (free tier, API key usage) on the main site with machine-readable pricing info.",
    "Offer a public status page and versioning to improve consistency perception.",
    "Expose a clear, structured error schema (e.g., JSON errors) and document common failure cases.",
    "Simplify initial setup: provide a hosted MCP endpoint or single command install to reduce first-try overhead."
  ]
}