Apis Guru
Server Details
APIs.guru MCP — keyless directory of 2,500+ public APIs and their OpenAPI specs. https://apis.guru
The last two health checks of this connector failed. If this server requires authentication, missing or expired test credentials may be the cause. A test profile lets Glama authenticate for health checks and discover tools; it is separate from your personal connections.
If you are the author, claim ownership, then add or update a test profile under Admin → Test Profile.
- Status
- Unhealthy
- Uptime
- 0.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP
- URL
- Repository
- pipeworx-io/mcp-apis-guru
- GitHub Stars
- 0
- Server Listing
- mcp-apis-guru
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 48 tools
The router family is a clear problem: ask_pipeworx, ask_pipeworx_beta (which the description admits is byte-identical to the stable router right now), ask_pipeworx_grounded, and deep_research all answer 'route my question to the catalog', leaving an agent to pick on vibes. The Polymarket cluster (polymarket_arbitrage vs polymarket_edges vs edge_tracker vs fill_risk vs kalshi_spread vs bet_research) and the row/claim trio (research_row, classify_row, validate_claim) overlap enough that selection requires reading long prose rather than recognizing a distinct purpose. Long descriptions mitigate some of this, but boundaries are prose-enforced, not structurally obvious.
Names are uniformly snake_case, which is a real plus, but the convention mixes verb_noun (search_apis, get_api, resolve_entity, validate_claim, compare_entities) with bare noun phrases (entity_profile, company_facts, recent_changes, prediction_calibration, resolution_audit, kalshi_weather_edge) and idiosyncratic forms (ask_pipeworx, discover_tools, suggest_questions). The _beta / _grounded suffixes on the ask_pipeworx family are the only structured variant marker, so the pattern is readable but not predictable.
48 tools for what is fundamentally a 'route questions to thousands of upstream data tools' server is heavy, and several slots are duplicates or near-duplicates (ask_pipeworx_beta, deep_research vs ask_pipeworx, polymarket_edges vs polymarket_arbitrage) rather than distinct capabilities. The breadth is defensible given the scope, but the surface has clearly outgrown what an agent can hold in working memory in one shot.
Coverage of the domain is genuinely broad and mostly closed-loop: identity resolution, financials, comparison, change feeds, prediction-market pricing/audit/fill-risk, memory (remember/recall/forget), and subscription lifecycle (subscribe/list/unsubscribe/recent_alerts). Notable gaps: fetch_url and the per-source primitives (polymarket_market, kalshi_market, deps.dev:version) are referenced as dependencies by research_row, scan_dependency, and resolution_audit but are not exposed as callable tools, creating dead ends when a lower-level lookup is needed. There is also no direct spec/OpenAPI fetch beyond get_api's URL listing.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
The world's public data, one API. License-cleared public data sources for agents, via MCP and REST.
HUD MCP — U.S. Department of Housing and Urban Development APIs.
MCP server for AI access to Swagger by SmartBear.
Swagger Petstore API (v1.0.27) as MCP for testing and prototyping powered by the HAPI MCP server
Related MCP Servers
- MIT
- AlicenseBqualityBmaintenanceTurn any OpenAPI 3.x, Swagger 2.0 or Postman spec into a curated MCP server84Apache 2.0
- AlicenseNot gradedqualityBmaintenanceMCP server providing token-efficient access to OpenAPI/Swagger specs via MCP Resources for client-side exploration.582 npm76MIT
- AlicenseNot gradedqualityDmaintenanceMCP server for querying Swagger/OpenAPI metadata efficiently from AI tools, enabling fast API discovery, search, and schema inspection.MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.