LinkFinder AI — OpenAPI Spec

Unofficial OpenAPI 3.0 spec for the LinkFinder AI API, built from their public docs.

Files

• openapi.yaml — the spec (source of truth)
• openapi.json — same spec, JSON format (some tools prefer this)

How to publish on GitHub

1. Push this repo (or add these files to an existing repo).
2. Get the raw file URL, e.g.: https://raw.githubusercontent.com/<your-username>/<repo>/main/openapi.json
3. Use that raw URL as the "spec URL" input in an API→MCP converter.

Notes on this spec

• LinkFinder AI uses a single dispatch endpoint (POST /) — the type field in the body selects the operation, rather than separate REST paths per feature. This spec models that as one path with an enum on type.
• linkedin_profile_to_linkedin_info always responds 202 (async) — poll /status/{job_id} until done.
• Any other endpoint (most commonly company_domain_to_employees) can also fall back to the same async 202/job_id shape if it runs past ~27 seconds — don't assume a fixed sync response shape purely from the endpoint name.
• Every request costs 1 credit, including failed/empty lookups.
• Auth: Authorization: Bearer YOUR_API_KEY on every request.

Updating

If LinkFinder AI adds endpoints or changes the type enum, update the enum list under components.schemas.EnrichmentRequest.properties.type in openapi.yaml, then re-run the conversion script below to regenerate openapi.json.

pip install pyyaml
python3 -c "
import yaml, json
with open('openapi.yaml') as f:
    data = yaml.safe_load(f)
with open('openapi.json', 'w') as f:
    json.dump(data, f, indent=2)
"