API Quality Check

Validate an OpenAPI spec, generate missing documentation, and produce test cases. Answers 'is my API spec production-ready?'

3 steps api-quality-check

What it runs

OpenAPI Validation
Source: Algorithmic (OpenAPI 3.x specification validation)
API Docs Generator
Source: Claude API (API documentation generation)
Test Case Generator
Source: Claude API (test case generation)

The answer returns every step’s result, so your agent can show or check each one.

Inputs

Field	Type	Required	What it is
`spec`	string	Yes	OpenAPI spec (JSON or YAML string)

How to call it

# Needs an API key. The whole check is charged once, at the listed price.
curl -X POST https://api.strale.io/v1/solutions/api-quality-check/execute \
  -H "Authorization: Bearer $STRALE_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
    "inputs": {
      "spec": "{\"openapi\":\"3.0.0\",\"info\":{\"title\":\"My API\",\"version\":\"1.0\"},\"paths\":{\"/users\":{\"get\":{\"summary\":\"List users\"}}}}"
    },
    "max_price_cents": 50
  }'

The bundle’s published example input.

Limitations

Validates against OpenAPI 3.0/3.1 specification — custom extensions and vendor-specific annotations are not checked
From step: OpenAPI Validation
What to do instead: For vendor extensions (x- prefixed), add your own validation rules in a post-processing step
Generated documentation reflects the provided API spec — undocumented behaviour and edge cases are not inferred
From step: API Docs Generator
What to do instead: Review generated docs against actual API behaviour and add examples for complex endpoints
Structured output is limited to about 2000 output tokens per call; requests exceeding this limit are refused rather than returned truncated. Requests with large results can exceed this limit, and repeating the same request does not help.
From step: API Docs Generator
What to do instead: Provide a smaller spec or split it into fewer endpoints per call
AI-generated test cases cover common paths — edge cases specific to your business logic may require manual test authoring
From step: Test Case Generator
What to do instead: Use generated tests as a starting coverage baseline and add domain-specific edge cases manually
Structured output is limited to about 2000 output tokens per call; requests exceeding this limit are refused rather than returned truncated. Requests with large results can exceed this limit, and repeating the same request does not help.
From step: Test Case Generator
What to do instead: Request fewer test cases per call, or set include_edge_cases to false