API Documentation Extractor

What does API Documentation Extractor — Structured API Intelligence do?

This Actor turns API documentation into structured, machine-readable API intelligence. Give it a documentation URL. It prefers an OpenAPI or Swagger specification when one exists, otherwise it extracts endpoints from HTML. It does not return scraped page text or raw HTML as the product.

Use it when you need a normalized JSON model of an API for developers, AI agents, integration platforms, or API research.

Why use it?

• Discover OpenAPI/Swagger specs automatically and treat them as the source of truth
• Extract HTTP endpoints, authentication, parameters, request bodies, and responses from HTML when no spec is published
• Capture webhooks, errors, rate limits, SDKs, and documented code examples when they appear in the docs
• Keep a sourceUrl on extracted entities so you can audit and debug results
• Run on Apify with scheduling, monitoring, and API/CLI access

This is an extractor, not a generic web scraper.

What data can it extract?

Field	Description
api	Name, description, version, base URLs, documentation URL
authentication	Bearer, API key, OAuth, Basic, HMAC, JWT when documented
endpoints	Method, path, parameters, request body, responses, source URL
webhooks	Event name, payload, signature verification when documented
errors	HTTP status, error code, message
rateLimits	Documented limits, or null if none are stated
sdks	Client libraries mentioned in the docs
codeExamples	Copy-only examples from the documentation
crawl	Pages visited, OpenAPI URL, whether a browser was used

How to extract API intelligence

1. Open the Actor in Apify Console.
2. Paste a documentation URL into Documentation URL (for example https://paystack.com/docs/api/).
3. Optionally lower Max pages for a cheaper run.
4. Click Start.
5. Download the dataset as JSON, or fetch OUTPUT from the default key-value store.

How much will it cost?

Compute cost scales with pages crawled. HTTP/Cheerio is the default and is inexpensive. A browser is used only when a page looks JavaScript-rendered and the initial HTML has no useful API content. Start with a smaller maxPages on large sites such as Stripe or GitHub.

Input

See the input tab for the full form.

• startUrl (required) — documentation URL
• maxPages — default 100
• includeCodeExamples — default true
• includeWebhooks — default true
• includeErrors — default true

Output

You can download the dataset as JSON, HTML, CSV, or Excel. The Actor writes one JSON object describing the API, and the same object to key-value store record OUTPUT.

{
  "api": {
    "name": "Payments API",
    "version": "1",
    "baseUrls": ["https://api.payments.test/v1"],
    "documentationUrl": "https://docs.payments.test/api"
  },
  "authentication": [{ "type": "bearer", "headers": ["Authorization"] }],
  "endpoints": [
    {
      "method": "POST",
      "path": "/transaction/initialize",
      "name": "Initialize Transaction",
      "authentication": { "type": "bearer" },
      "requestBody": {
        "email": { "type": "string", "required": true },
        "amount": { "type": "integer", "required": true }
      },
      "responses": { "200": { "description": "Successful response" } },
      "sourceUrl": "https://docs.payments.test/api"
    }
  ],
  "webhooks": [],
  "errors": [],
  "rateLimits": null,
  "sdks": [],
  "codeExamples": [],
  "crawl": { "pagesVisited": 1, "pagesProcessed": 1, "durationMs": 1200, "sourceUrls": [] }
}

Supported documentation types

• OpenAPI 3 and Swagger 2 (openapi.json, swagger.yaml, /api-docs, and links from the docs)
• Static HTML API reference pages
• JavaScript-rendered docs, via a Playwright retry for pages whose initial HTML has no useful content

Limitations

• The Actor does not invent endpoints, webhooks, or auth schemes. Missing documentation stays missing (unknown or empty arrays / null).
• It does not generate client code. Examples are copied only when the docs include them.
• Selectors are provider-agnostic. Unusual one-off doc layouts may yield fewer fields.
• Change detection, API comparison, and historical versions are out of scope.

API usage example

apify call api-documentation-extractor --input '{
  "startUrl": "https://paystack.com/docs/api/",
  "maxPages": 40,
  "includeWebhooks": true,
  "includeErrors": true
}'

Use the Apify API to read dataset items after the run, or the OUTPUT record in the default key-value store.

FAQ

Our Actors extract publicly documented API information. Do not use this Actor to bypass authentication, scrape private user data, or violate a site's terms of service. If your results could contain personal data, GDPR and similar laws may apply.

Open issues on the Issues tab. Programmatic runs are documented on the API tab.