Scrape a single page
Pass a URL, get back one page of clean markdown. Headers, lists, code blocks, and tables preserved.
Documentation
Introduction
Syttra turns any public webpage into clean, AI-ready content. Give it a URL, get back markdown, plain text, or structured data — ready to feed your LLM pipeline or RAG index.
Private beta: the API is running for beta users only. Request access via hello@syttra.com.
What you can do
Crawl a full site
Follow internal links up to a depth limit. Same-domain only. robots.txt and ToS enforced.
Export in the format you need
Markdown or plain text today. HTML, PDF, and Word in v1.1.
Track job progress
Async-first: create a job, poll status, download results when ready.
How it works
- 1
Authenticate
Every request carries an API key in the
Authorization: Bearer sk_live_...header. Keys are per-account, rotatable, and hashed at rest. - 2
Create a job
POST /v1/jobswith your URL and options. Returns202 Acceptedwith a job ID — Syttra queues the work, you move on. - 3
Poll for status
GET /v1/jobs/{id}returns the job state:queued→running→completed(orfailed). - 4
Download results
GET /v1/jobs/{id}/resultstreams the cleaned content back. Cached for 24h, then deleted.