Inspect an OpenAPI document
Read an OpenAPI 3.0, 3.1 or Swagger 2.0 document and report its structure: title, servers, paths and methods, schemas, tags and a bounded reference audit that n…
Working…
Questions people ask
- Which specification versions are supported?
- Swagger 2.0 and OpenAPI 3.0 and 3.1 are supported, read only. Anything else is refused by name: the message repeats the version the document declares, so a 2.1 claim or a missing version field is immediately visible instead of producing a half inspection.
- What happens to external references?
- They are counted and named, and that is all. A $ref pointing at another file or an http:// URL is never fetched, never contacted; the audit stays strictly local to the bytes you gave. Unresolved local references are counted too, with the first few named by where they sit in the document.
- What does the reference audit check locally?
- Every #/ reference is walked against the document root like a JSON pointer, so a local reference that points at a missing path is counted as unresolved and named. The walk itself is depth-capped and cycle-safe, so an aliased or pathological document cannot make the audit run away.
- Does an unresolved reference mean my document is broken?
- It means this tool could not walk that pointer inside the document, and the report names it precisely. It does not mean the document is invalid overall; external tooling may resolve what this local audit cannot. There is no conformance claim here, deliberately.
- What do I get as output?
- Two artifacts: a readable report.txt with the structure laid out in sections, and report.json with the same numbers in machine-readable form for pipelines and checklists.
- What happens to my files?
- The upload is removed as soon as processing finishes; the reports stay downloadable for about 10 minutes, then they are deleted automatically.