Vercelvercel.com

ScoredAug 12, 2026, 07:56 PM

Vercel's API program shows strong foundational API design, with a clean, well-typed contract that partners and agents can build against confidently. The areas most in need of attention are agent understanding and agent discovery: partners' AI agents currently lack the structured, machine-readable context they need to interpret and act on the API surface without guessing, and the documentation is not surfaced in ways that let agents — or the partners directing them — find and consume it autonomously.

API DesignA clean, typed, well-governed API contract agents can reason about4 pass1 warn0 fail100A+
SignalPointsFindingsRationale
passMachine-readable, versioned contractvia spec25/25The API declares version "0.0.1" and uses url-based versioning, so an agent can pin to a stable contract. Investigated: spec 100%, docs 100%, cli 50%.A current OpenAPI version with a declared versioning scheme lets agents reason about the contract.
passAuth declared & discoverablevia docs25/25authentication is documented on the docs site at https://vercel.com/login?next=%2Fauthentication, even though it isn't declared in the API spec. Investigated: docs 100%, wellknown 100%, spec 50%, cli 0%.Agents can only authenticate when auth is declared, scoped, and discoverable.
warnSchema coverage & depthvia sdk24.9/25734/735 (100%) public params are fully typed (no any/**kwargs). Analyzer reports not all methods have a derivable input schema. Investigated: sdk 100%, spec 67%.Typed, complete request/response schemas are what make agent function-calling possible.
passSecurity & governance hygienevia spec15/15No credential-shaped strings detected in spec. Investigated: spec 100%, sdk 100%, wellknown 100%.No leaked secrets, no critical lint violations, no OWASP API Top-10 spec smells, and a published vulnerability-disclosure channel.
passExample coveragevia spec10/10Only 51% of parameters and responses (888 of 1758) include examples (target 80%+), so an agent guesses payload shapes the schema under-specifies. Investigated: spec 100%, docs 100%, sdk 100%, cli 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 examples5 pass0 warn0 fail100A+
SignalPointsFindingsRationale
passSelf-service developer portalvia docs29/29Signup page at https://vercel.com/signup (client-rendered, so its contents couldn't be classified); free tier or sandbox is documented in pricing/docs. Investigated: docs 100%.Self-serve key/account creation is the fast first call for partners, with no sales gate.
passQuickstart presentvia docs25/25Quickstart at https://vercel.com/docs/getting-started-with-vercel has a runnable code sample in its plain-text .md variant (https://vercel.com/docs/getting-started-with-vercel.md). Investigated: docs 100%.A quickstart is the fastest path from landing page to first successful call.
passCode samples in docsvia docs18/18Docs show code samples in 3 language(s) (top: typescript, curl, javascript), and 5 of 20 sampled pages offer ≥ 2 languages side by side. Investigated: docs 100%.Multi-language samples shorten time-to-first-call.
passDescription completenessvia spec15/15Only 83% of operations and parameters have substantive descriptions (326 of 377 ops, 1051 of 1324 params; target 90%+). Investigated: spec 100%, sdk 95%.Complete descriptions are the context humans and agents need to use endpoints.
passChangelog publishedvia spec13/13changelog is documented on the docs site at https://vercel.com/docs/cli/release-notes, even though it isn't declared in the API spec. Investigated: spec 100%, docs 100%, sdk 100%.A published changelog lets partners track changes without surprise.
Agent DiscoveryPartners and their agents can find your APIs — llms.txt, registries, crawlable and reachable docs2 pass2 warn0 fail82A
SignalPointsFindingsRationale
passRegistry & SDK presencevia docs28/28Indexed on Context7 (websites/vercel, 14255 snippets). Investigated: docs 100%, mcp 100%, sdk 83%, cli 0%, wellknown 0%.Listing in MCP registries and publishing SDKs puts the API where agents and their tooling look.
warnDocs reachable, not hard auth-gatedvia docs22.5/301 of 50 sampled pages return 200 for non-existent URLs (soft 404). Investigated: docs 75%.Agents can only index and fetch docs they can reach — past auth gates and over correct HTTP semantics.
warnllms.txt present, valid & comprehensivevia docs16.9/30No llms.txt directive found in HTML of any of 50 sampled pages. Investigated: docs 56%, sdk 0%.A valid, comprehensive llms.txt is the machine-readable entry point for agents.
passCrawlable / AEOvia wellknown12/12robots.txt lets all monitored AI crawlers reach the docs paths. 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 docs3 pass3 warn0 fail83A
SignalPointsFindingsRationale
passMachine-readable errors (RFC 9457)via docs28/28Documents 21 distinct 4xx/5xx error response codes, so an agent can branch on specific failures instead of guessing. Investigated: docs 100%, sdk 100%, spec 50%, cli 0%.RFC 9457 problem details and a documented error-code inventory let agents parse failures without burning tokens.
passOperation purpose clarityvia spec25/2597% of operations (367 of 377) have both a clear summary and a descriptive name, so an agent can pick the right endpoint. Investigated: spec 100%, sdk 100%.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 docs13.4/2220 of 50 pages have substantive content differences between markdown and HTML (avg 30% missing). Investigated: docs 61%.Server-rendered, clean, small-footprint docs are what an agent can cheaply fetch and parse correctly.
passDocs structured datavia docs8/8Machine-readable JSON-LD Article markup on 12 of 20 assessed pages (60%), with dateModified present. Investigated: docs 100%.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.
warnAgent instructions file (AGENTS.md)via sdk5/10No agents.md or skill.md context file found (nice-to-have for agent operation). Investigated: sdk 50%, cli 0%, 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.
warnDescription consistency across surfaces1/7Mean pairwise description similarity across 3 surfaces (spec, docs, sdk) is 5% (threshold 35% for full credit).Every surface tells the same story about what the product is.
Agent UsabilityAgents have the context to use your APIs reliably, not just find them5 pass0 warn0 fail100A+
SignalPointsFindingsRationale
passIdempotency documentedvia sdk27/27Built-in retry machinery is present (grep-derived). Investigated: sdk 100%, spec 0%, docs 0%, cli 0%.Documented idempotency lets agents retry safely.
passRate-limit signalingvia spec22/22rate-limit behavior is documented on the docs site at https://vercel.com/docs/limits, even though it isn't declared in the API spec. Investigated: spec 100%, docs 100%.Machine-readable rate-limit headers let agents throttle adaptively.
passPagination documented & consistentvia spec22/22Pagination parameters are present on the 16 list endpoint(s), so an agent can page through results. Investigated: spec 100%, docs 100%, sdk 20%.Consistent, documented pagination lets agents traverse collections.
passSandbox separationvia spec20/20sandbox is documented on the docs site at https://vercel.com/docs/sandbox, even though it isn't declared in the API spec. Investigated: spec 100%, docs 100%, sdk 100%.An isolated environment lets agents exercise destructive operations safely.
passRunnable collection with test scriptsvia sdk9/9README is present and includes a runnable quickstart. Investigated: sdk 100%, platform 0%.A public, maintained collection with assertions is runnable truth agents validate against.
Resources Discovered

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

Agent hintsContext7 (14,255)
APIs analyzed1Vercel API
Want to improve your results?