One API key unlocks scraping, extraction, screenshots, sitemaps, classification, and semantic search. No separate services, no stitching together five vendors.
START FOR FREE →Single URL → clean markdown. Images stripped, boilerplate removed, structure preserved.
Multi-page crawl — discover and scrape all pages on a site, return clean markdown.
Full AI business intelligence — classification, knowledge items, and insights.
Headless browser screenshot of any URL, returned as a base64 PNG.
Discover every public URL on a domain — crawl the whole site in one call.
Hybrid semantic search over everything you've indexed. Results in under 300ms.
A single knowledgesdk_live_* key unlocks every endpoint. No per-service auth, no separate tokens.
Every response follows the same shape. No surprises, no version negotiation.
Run /v1/business synchronously or use /v1/business/async with a webhook for long jobs.
Register a callbackUrl and we'll POST results when async jobs finish. No polling needed.
Sign up and receive a knowledgesdk_live_* key instantly. No credit card for the free tier.
POST to any endpoint with your key in the x-api-key header. That's all the setup there is.
Sitemap → Extract → Search. Chain our endpoints to build full RAG pipelines in minutes.
Rate limits, usage tracking, and async jobs are all included. No extra infrastructure needed.
Chain our endpoints to build a complete RAG pipeline. Discover pages with the sitemap endpoint, extract and index them, then search in natural language.
One base URL. One API key. The same auth header on every request.
STEP 1 — DISCOVER ALL PAGES
const { urls } = await client.sitemap({ url: "https://docs.stripe.com" });
STEP 2 — EXTRACT & INDEX EACH PAGE
for (const url of urls) { await client.extract({ url, store: true }); }
STEP 3 — SEARCH IN PLAIN ENGLISH
const { results } = await client.search({ query: "how do I handle webhook retries?", limit: 5, });