Utilities
KnowledgeSDK provides a set of low-level utility endpoints that give you direct access to individual capabilities of the platform. These are useful when you need raw data from a single step of the pipeline rather than the full business extraction workflow.
These are low-level tools. Most users should use /v1/business to automatically extract, classify, index, and make content searchable in a single API call. Use these utilities only when you need fine-grained control over individual operations.
Available utilities
Fetch any URL and return clean, LLM-ready markdown. Handles JavaScript rendering, anti-bot bypasses, and HTML cleanup automatically.
Capture a full-page screenshot of any URL and receive it as a base64-encoded PNG image.
Discover all URLs on a website by crawling its sitemap. Supports sitemap indexes, WordPress sitemaps, and automatic www/non-www fallback.
Looking for business classification? The /v1/classify endpoint has been removed. Classification is now included as part of /v1/business.
Authentication
All utility endpoints require an API key passed via the x-api-key header. Your key must start with sk_ks_.
curl -X POST https://api.knowledgesdk.com/v1/extract \
-H "x-api-key: sk_ks_your_api_key" \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com"}'Rate limits
Each utility endpoint has its own rate limit to ensure fair usage:
| Endpoint | Rate limit |
|---|---|
/v1/extract | 30 requests / minute |
/v1/screenshot | 20 requests / minute |
/v1/sitemap | 30 requests / minute |
If you exceed the rate limit, the API returns a 429 Too Many Requests response. Implement exponential backoff in your retry logic, or use the official SDKs which handle this automatically.