Coherecohere.com

ScoredJul 24, 2026, 08:33 AM

Cohere's API program shows strong fundamentals in API design, agent usability, and discovery — partners and agents can find and run against a well-structured contract. The two areas that need priority attention are agent understanding and developer onboarding: there are no structured cues or instruction files to help AI agents parse and act on Cohere's docs autonomously, and partners lack the quickstart resources and embedded code samples needed to reach a first successful call quickly.

API DesignA clean, typed, well-governed API contract agents can reason about3 pass2 warn0 fail94A+
SignalPointsFindingsRationale
passAuth declared & discoverablevia spec25/25Authentication required (global or on every operation). Investigated: spec 100%, docs 100%.Agents can only authenticate when auth is declared, scoped, and discoverable.
warnSchema coverage & depthvia sdk24.9/251363/1377 (99%) public params are fully typed (no any/**kwargs). Analyzer reports not all methods have a derivable input schema. Investigated: sdk 99%, spec 80%.Typed, complete request/response schemas are what make agent function-calling possible.
warnMachine-readable, versioned contractvia spec18.8/25info.version="1.0.0" set but no versioning scheme (URL / header / media type) detected. Investigated: spec 75%, docs 50%.A current OpenAPI version with a declared versioning scheme lets agents reason about the contract.
passSecurity & governance hygienevia spec15/15No credential-shaped strings detected in spec. Investigated: spec 100%, wellknown 0%.No leaked secrets, no critical lint violations, no OWASP API Top-10 spec smells, and a published vulnerability-disclosure channel.
passExample coveragevia spec10/108% example coverage. Investigated: spec 100%, sdk 100%, docs 0%.Examples carry shape semantics schemas under-specify — for humans and agents alike.
Developer ExperienceThe context both developers and agents need to integrate fast — onboarding, code samples, complete descriptions and worked examples3 pass0 warn2 fail66C
SignalPointsFindingsRationale
passSelf-service developer portalvia docs29/29Self-service signup at https://dashboard.cohere.com/welcome/register; no free tier or sandbox language detected. Investigated: docs 100%.Self-serve key/account creation is the fast first call for partners, with no sales gate.
passDescription completenessvia spec15/1585% description completeness (target 90%+). Investigated: spec 100%, sdk 76%.Complete descriptions are the context humans and agents need to use endpoints.
passChangelog publishedvia spec13/13changelog documented in the docs site at https://docs.cohere.com/v2/changelog. Investigated: spec 100%, docs 100%, sdk 100%.A published changelog lets partners track changes without surprise.
failQuickstart presentvia docs0/25No quickstart/getting-started page found at the conventional paths. Investigated: docs 0%.A quickstart is the fastest path from landing page to first successful call.
failCode samples in docsvia docs0/18No detectable code samples across 2 sampled docs pages. Investigated: docs 0%.Multi-language samples shorten time-to-first-call.
Agent DiscoveryPartners and their agents can find your APIs — llms.txt, registries, crawlable and reachable docs3 pass1 warn0 fail95A+
SignalPointsFindingsRationale
passDocs reachable, not hard auth-gatedvia docs30/30All 50 sampled pages are publicly accessible. Investigated: docs 100%.Agents can only index and fetch docs they can reach — past auth gates and over correct HTTP semantics.
passRegistry & SDK presencevia docs28/28Indexed on Context7 (websites/cohere, 4449 snippets). Investigated: docs 100%, sdk 58%, cli 0%, mcp 0%, wellknown 0%.Listing in MCP registries and publishing SDKs puts the API where agents and their tooling look.
warnllms.txt present, valid & comprehensivevia docs24.4/30llms.txt covers 189/278 sitemap doc pages (68%); 89 missing. Investigated: docs 81%, sdk 0%.A valid, comprehensive llms.txt is the machine-readable entry point for agents.
passCrawlable / AEOvia wellknown12/12Docs paths crawlable by all monitored AI agents. Investigated: wellknown 100%.Bots allowed plus a fresh sitemap make docs findable by agent crawlers.
Agent UnderstandingAgents can correctly interpret your APIs — machine-readable errors, consistent descriptions, structured data, parseable docs1 pass3 warn2 fail72B
SignalPointsFindingsRationale
passMachine-readable errors (RFC 9457)via docs28/28No 4xx/5xx response codes (or default error response) documented anywhere in the spec. Investigated: docs 100%, sdk 100%, spec 50%.RFC 9457 problem details and a documented error-code inventory let agents parse failures without burning tokens.
warnOperation purpose clarityvia spec22.5/2575% of operations are agent-inferable (target 90%+). Investigated: spec 90%.Agents select the right endpoint from its summary + operationId; clear, named operations make tool-selection reliable — the strongest driver of correct tool choice.
warnAgent-navigable, token-efficient docsvia docs13.2/2212 of 50 pages have substantive content differences between markdown and HTML (avg 12% missing). Investigated: docs 60%.Server-rendered, clean, small-footprint docs are what an agent can cheaply fetch and parse correctly.
warnDescription consistency across surfaces1.5/7Mean pairwise description similarity across 2 surfaces (spec, sdk) is 8% (threshold 35% for full credit).Every surface tells the same story about what the product is.
failAgent instructions file (AGENTS.md)via wellknown0/10No AGENTS.md at the site root or /.well-known/. Investigated: wellknown 0%.An AGENTS.md gives coding agents explicit setup, auth, and usage instructions to interpret and operate the API — beyond llms.txt's link index.
failDocs structured datavia docs0/8Neither JSON-LD nor OpenGraph/meta tags detected across 2 sampled pages — likely a CSR-only docs site. Investigated: docs 0%.Structured data (JSON-LD/schema.org) on docs pages gives agents an unambiguous parse target and is what answer engines cite. Detected on the JS-rendered head (Firecrawl) for a bounded page budget, so JS-injected JSON-LD is now caught; pages we can't render are excluded rather than failed.
Agent UsabilityAgents have the context to use your APIs reliably, not just find them3 pass1 warn0 fail95A+
SignalPointsFindingsRationale
passIdempotency documentedvia sdk27/27Built-in retry machinery is present (grep-derived). Investigated: sdk 100%, spec 0%, docs 0%.Documented idempotency lets agents retry safely.
passRate-limit signalingvia spec22/22rate-limit behavior documented in the docs site at https://docs.cohere.com/docs/rate-limits. Investigated: spec 100%, docs 100%.Machine-readable rate-limit headers let agents throttle adaptively.
passSandbox separationvia sdk20/20Sandbox environments are exposed (sandbox). Investigated: sdk 100%, spec 0%, docs 0%.An isolated environment lets agents exercise destructive operations safely.
warnRunnable collection with test scriptsvia platform4.5/9No test scripts found in the workspace's collections. Investigated: platform 50%.A public, maintained collection with assertions is runnable truth agents validate against.
naPagination documented & consistent—/22No surface produced evidence for this capability in this run.Consistent, documented pagination lets agents traverse collections.
Resources Discovered

The public resources we found for Cohere — the evidence behind the score. All discovered from public sources; nothing here requires access to your systems.

Agent hintsContext7 (4,449)
APIs analyzed5API Reference, 4_Misc, 1_Chat, 2_Embed, 3_Rerank
Want to improve your results?