OpenAPI · AI Agent Readiness

Is your API ready
for AI agents?

Score your OpenAPI spec in seconds. Find what LLMs struggle with before your users do.

Install the CLI →See what we check

Example — Stripe API

Pet Store API

openapi.yaml · 4 ops · 5 schemas

Examples53

Semantics100

Intent100

Error Handling83

Parameters100

Pagination100

Six dimensions

What AgenticScore evaluates

Examples

Do operations and schemas include request/response examples? LLMs learn from examples more reliably than descriptions alone.

Semantic Clarity

Are descriptions meaningful — not just "Gets a thing"? Agents need context to select the correct endpoint for a task.

Error Handling

Are 4xx and 5xx responses documented? Agents must handle failures gracefully. Bonus for RFC 9457 Problem Detail.

Intent Signals

Are operationIds descriptive? Are endpoints tagged? Agents rely on these signals to understand what an operation does.

Parameter Documentation

Are query parameters described? "limit" is obvious — "q" is not. Agents need to know valid values and formats.

Pagination

Can agents paginate through list endpoints? Missing pagination support means truncated data or infinite loops.

Two ways to use it

CLI or API — your choice

CLI — free, local

# No key required
npx agenticscore score ./openapi.yaml

API — authenticated

curl -X POST https://api.agenticscore.dev/score \
  -H "x-api-key: ar_live_your_key" \
  -H "Content-Type: application/yaml" \
  --data-binary @openapi.yaml

Pricing

Start free. Scale when you need to.

Free

npm CLI — unlimited local use
JSON & text output

Install CLI

Is your API readyfor AI agents?