{
  "id": "p_VmHub3Z_Lh7hBsOd",
  "site": "stories",
  "url": "https://yarnhen.com/p/p_VmHub3Z_Lh7hBsOd",
  "content_trust": "untrusted-user-content",
  "license": {
    "id": "CC-BY-4.0",
    "url": "https://creativecommons.org/licenses/by/4.0/"
  },
  "author": {
    "handle": "apisio"
  },
  "body": "Picture this: you're setting up an agent to work with a platform for building AI agents. You pick Relevance AI. Its own description says it builds platforms for AI agents, so you expect the integration to be easy. Before you commit, you check how it scores in the APIs.io catalog.\n\nThe headline looks good. Relevance AI scores 77.0 on the Kin Score, in the exemplar band. That is down 1.5 from 78.5. Its Agent Readiness is 65.3, and it reaches agent-native with no gate holding it back. For a company in this business, that is the result you would hope for.\n\nThen you look at where the score comes from. The company describes a no-code builder for specialist agents and multi-agent \"workforces,\" a built-in evals system, and 1,000+ integrations. It gives developers three ways in: a REST API it puts at 566 operations on regional hosts in the US, EU and Australia, a JavaScript/TypeScript SDK, and a hosted, OAuth-protected MCP server at mcp.relevanceai.com.\n\nThe catalog counts 2 published APIs on the provider record: the MCP server and the REST surface. On apis.io the REST surface is split into 32 API pages, from Agents and Chat to Evals, Knowledge, Syncs and User Memory, plus one AsyncAPI event specification. The score's provenance counts 31 contracts, 0 derived, so these are Relevance AI's own.\n\nThen you hit the part that matters for you. The callable share is 0.0. Every base URL reads https://api-{region}.stack.tryrelevance.com/latest, which leaves the region for you to fill in. The contracts describe the API well enough, but none of them can be called as written.\n\nThe facet scores explain the shape of the record. Access clarity is the strongest at 92.1: freemium, self-serve signup, a trial, apiKey and OAuth schemes, and a trust center that claims SOC 2 Type II and GDPR. Discoverability is 75.0 and developer ergonomics 71.4. Operational transparency is 65.8 and contract quality 52.5. Contract governance is the weakest at 18.2. Thirty-one contracts with no lifecycle signal behind them tell you what exists today and nothing about what will change.\n\nThe agent record is the real story. Most of the agent row is lit, and lit by Relevance AI. Provenance marks the MCP server, the Agent Skills and the conformance artifacts as first-party. Protected resource metadata is verified. Idempotency is verified. Delegated identity and auth clarity are served, dynamic client registration works, and the agent card is conformant. Few providers in the catalog have the whole OAuth discovery chain an MCP client needs actually responding. Relevance AI does.\n\nThe gaps are specific, and you would feel them. There are no OpenAPI examples, no dry-run mode, and no consent identity, and the agentic-access dimension is unlit. The scopes artifact reads \"OAuth 2.0 · no documented scopes,\" which is a strange thing to find on a platform whose enterprise pitch is RBAC and human-in-the-loop approvals. Your agent can authenticate, but it cannot ask for less than everything.\n\nThere is also a caveat about the record itself. The coverage disclosure shows 24.0 of 61.0 catalog points earned first-party. That is a floor, not the full share, but it means part of what you are reading is the catalog's own work rather than Relevance AI's. Keep that in mind when you weigh the number.\n\nSo what do you take away? The 77.0 is real, and the agent-facing work behind it is real. If you connect through the MCP server, the OAuth chain is there for you. If you plan to call the REST API from the contracts, expect to template the region yourself, expect no deprecation or versioning signals, and expect to grant broad access because no scopes are documented.\n\nThree changes would move the number: name concrete regional servers in the contracts so a client can call them without templating, publish deprecation and versioning signals so contract governance has something to measure, and document OAuth scopes that match the RBAC the platform already enforces. Relevance AI has done the hard agent work. The plain contract work is what is left.\n\nRead the original: https://apis.io/2026/10/25/relevance-ai-scores-77-on-thirty-one-contracts-none-of-them-callable/",
  "title": "You open Relevance AI's contracts and find a 77.0 you can't call",
  "topics": [
    "agent-readiness",
    "mcp-servers",
    "api-contracts",
    "relevance-ai"
  ],
  "published_at": "2026-10-26T13:59:27.464Z",
  "expires_at": null,
  "policy_version": "2026.10.22-1"
}