◎ Discry Score
anthropic.com
ai · API
B
0 / 100
DISCOVERY0
COMPREHENSION0
Category leader: 99 (A)
Discry your API →
AI · RANK #32 OF 43

Anthropic is Good to agents.

Discry independently scored how well an AI agent can discover and understand the Anthropic API from what’s public — not whether it’s usable. Below: every signal we checked, what’s costing the score, and what to change.

Discry your API — freeView the docs ↗

SCORED UNDER RUBRIC 1.2 · A full re-launch under Discry Score 2.5 — a new behavioral instrument, not comparable to these scores — is in progress.

Discovery

45% of score · 52/100
OpenAPI specA machine-readable OpenAPI/Swagger spec agents can parse.Fail
llms.txtAn llms.txt index that points agents to the docs that matter.Pass
llms.txt qualityThe llms.txt is focused, current, and well under the size budget.Partial
llms-full.txtA full-text bundle agents can load in one request.Pass
AGENTS.mdAn AGENTS.md that tells coding agents how to build on the API.Fail
.well-known/mcp.jsonA discoverable MCP manifest at a well-known path.Fail
MCP registryThe API is listed in a public MCP registry.Pass
robots.txt AI directivesrobots.txt allows (or explicitly guides) AI crawlers.Pass
SitemapA sitemap so agents can enumerate the docs surface.Pass

Comprehension

55% of score · 100/100
Task-oriented descriptionsEndpoints described by what they accomplish, not just their shape.Pass
Realistic examplesRunnable, real-world request/response examples.Pass
Multi-step workflowsDocs that chain calls into complete jobs an agent can follow.Pass
Error-recovery guidanceDocumented failure modes and how to recover from them.Pass
Answer-first formatThe answer leads; preamble does not bury it.Pass
Capability boundariesClear limits — what the API can and cannot do.Pass
Naming consistencyConsistent, predictable naming across endpoints.Pass
Heading hierarchyClean heading structure agents can navigate.Pass
Markdown docsDocs available as clean markdown, not JS-rendered HTML only.Pass
Token efficiencyDocs are concise enough to fit an agent context window.Pass

What we found

  • Anthropic's comprehension is flawless (100/100): all 10 checks pass including the new multiStepWorkflows (tool use guides, agent building guides) and errorRecoveryGuidance (typed error responses with actionable recovery steps)
  • The llms.txt indexes 1,284 pages but is a broad site-wide index rather than an API-focused right-sized file — it earns a partial on the new llmsTxtQuality check despite being present and well-structured
  • The biggest discovery gap remains the absence of a public OpenAPI spec (weight 5) — only unofficial community-maintained specs exist, which is notable for a company whose API is widely consumed by agents
  • No .well-known/mcp.json or AGENTS.md exists despite Anthropic creating the MCP protocol — two discovery signals at zero from the protocol's creator
  • llms-full.txt at 57MB+ provides exhaustive markdown coverage of the entire documentation surface — an agent can ingest the complete API without scraping

What to change

Prioritized by impact on discoverability. You (or your docs platform) deploy these — Discry never touches your API.

  1. 01Publish an official OpenAPI specification at docs.anthropic.com/openapi.json — this single change would add 5 weighted points and lift discovery from 56 to 75
  2. 02Add .well-known/mcp.json with tool declarations for Messages, Batches, and Token Counting APIs — leading by example as the MCP creator
  3. 03Create a focused API-only llms.txt (<50KB) covering core API endpoints, auth, and capabilities — keep the current 1,284-page index as a separate resource
  4. 04Add AGENTS.md to anthropics/anthropic-sdk-python and anthropics/anthropic-sdk-typescript with API coding patterns and common pitfalls
  5. 05Document idempotency support for the Messages API to help agents safely retry requests

Execution coverage · INFORMATIONAL, UNSCORED

Whether an agent can actually complete a call and recover from errors is the deeper Audit layer — documented here, but not part of the Discry Score.

api_keyoauth2_workload_identity_federation✓ Error format documented✓ Rate limits documented✓ Pagination documented✕ Idempotency documented

API key via x-api-key header plus OAuth2 via Workload Identity Federation. JSON error responses with typed error codes and request IDs. Detailed rate limits per model/tier with anthropic-ratelimit-* response headers. Cursor-based pagination. No idempotency key support documented.

See your own Discry Score.

Drop your API docs URL. See what an agent sees — in 60 seconds, free.

Discry your API — free