Guides
Task-oriented walkthroughs of authentication, scraping, errors, limits, monitors, screenshots, search, and webhooks.
For AI agents: https://bytekit.com/llms.txt — every docs page is available as markdown by appending .md
Task-oriented walkthroughs of the ByteKit API. Start with Choosing an endpoint to find the endpoint that fits the job in front of you, then follow its guide. Every request carries an API key; Authentication covers getting one onto a request. Per-endpoint request and response shapes live in the API Reference.
Choosing an endpoint
Which endpoint to use when two look alike — scrape or fetch, the three bulk APIs, schema or scrape, the monitor types, and search or sitemap.
Authentication
How to authenticate with the ByteKit API using API keys.
Scraping
Scrape any URL as HTML, markdown, or structured content — formats, options, latency, and async polling.
Fetch
Fetch the raw response from any URL — the X-Fetch-* headers, options, cost, and errors.
Errors
ByteKit error response shape, common error codes, and retry guidance.
Rate Limits
How ByteKit enforces rate limits, concurrency slots, and balance checks.
Monitors
Detect page changes on any URL and receive webhook notifications.
Screenshots
Capture a screenshot of any URL — options, response, cost, and errors.
Search
Search the web — options, response, cost in credits, and errors.
Schema extraction
Turn a page into structured JSON that follows your schema — options, response, cost, and errors.
Bulk
Submit many URLs in one request — which bulk API to use, results by webhook or polling, cancelling, cost, and limits.
Sitemap
Discover the URLs on a domain — start a job, read its status, download the results file, get a webhook, and what it costs.
Webhooks
Verify webhook deliveries — signature and timestamp headers, the HMAC scheme, and verification code.
MCP Server
Call ByteKit web capture as a native tool from Claude, Cursor, and any MCP-compatible agent — hosted over HTTP or run locally over stdio.
Choosing an endpoint
Pick the right endpoint when two look alike — scrape or fetch, which bulk API, schema or scrape, which monitor type, and search or sitemap.