◎ Discry Score
cohere.com
ai · API
B
0 / 100
DISCOVERY0
COMPREHENSION0
Category leader: 99 (A)
Discry your API →
AI · RANK #22 OF 43

Cohere is Good to agents.

Discry independently scored how well an AI agent can discover and understand the Cohere 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 · 76/100
OpenAPI specA machine-readable OpenAPI/Swagger spec agents can parse.Pass
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.Pass
llms-full.txtA full-text bundle agents can load in one request.Partial
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.Partial
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

  • An agent looking for Cohere finds it instantly and parses it cleanly: a root llms.txt with explicit AI-agent instructions, .md append on every page, per-section /llms.txt indexes, and a hosted docs MCP server (docs.cohere.com/_mcp/server) make these docs unusually agent-native.
  • Comprehension is exemplary — task-oriented endpoint descriptions, four-language realistic code examples (Python/TypeScript/Java/Go), consistent snake_case fields, and granular per-model rate-limit tables mean an agent can call the API with almost no guesswork.
  • Error handling is genuinely agent-usable: every status code (400/401/402/404/429/499/500) pairs an explicit resolution step with specific, often self-correcting example messages (e.g. "setting prompt_truncation='AUTO'" or "add or update your payment method" with a direct link).
  • The single biggest discovery gap is AGENTS.md — absent from both the docs repo (cohere-developer-experience) and the SDK repos (cohere-python) — so a coding agent that clones the SDK gets no repo-level orientation, and it costs the most discovery weight.
  • Standard MCP discovery is only partial: there is no /.well-known/mcp.json despite a working docs MCP server, and Cohere shows up in MCP registries mainly through third-party integrations rather than an official first-party listing.

What to change

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

  1. 01Add an AGENTS.md to cohere-python (and the docs monorepo) — the highest-impact discovery fix (weight 3): orient coding agents on installation, the CO_API_KEY env var, and the ClientV2() entry point.
  2. 02Publish /.well-known/mcp.json pointing at the existing docs MCP server (docs.cohere.com/_mcp/server) so agents can auto-discover it via the standard well-known path instead of only finding it inside llms.txt.
  3. 03Turn the root /llms-full.txt into a true comprehensive markdown dump — it currently mirrors the tiny llms.txt index — so an agent can pull full API context in a single fetch.
  4. 04Get an official Cohere MCP server listed in Glama, Smithery, or PulseMCP to convert the partial registry signal into a pass.

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 key (Bearer token)OpenAI SDK compatibility layer✓ Error format documented✓ Rate limits documented✕ Pagination documented✕ Idempotency documented

Cohere documents API-key (Bearer) authentication plus an OpenAI-SDK compatibility layer, a machine-parseable JSON error format with per-status-code descriptions and specific example messages, and granular per-model and per-endpoint rate-limit tables (trial vs production). Pagination is not prominently documented for the core endpoints (chat/rerank/embed do not paginate) and idempotency keys are not mentioned.

See your own Discry Score.

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

Discry your API — free