Crawl the web
with your agent, pay per call.
Enter a URL and fetch the data. Your agents crawl web pages, extract rich metadata, and build datasets — without you managing scraping infrastructure. Billed per call.
Built for teams that need web data without the scraping stack.
Four shapes of customer who benefit most. If you recognize your workflow, the per-call model is in your favor.
Extract rich metadata from any page.
Pull title, meta description, canonical URL, Open Graph tags, and heading structure from any page in one call. Perfect for content audits and competitive analysis — feed a list of URLs in, get structured metadata out, with no scraping pipeline to maintain.
Makes web pages simple to access.
Any web page becomes accessible as a single tool call — the agent fetches the URL, gets back clean text ready for embedding, and moves on. Ship agents faster with competitive pay-per-fetch micropayments. No proxy management, no rendering pipeline, no rate-limit negotiation.
Build training datasets at scale.
Crawl public web content to build training datasets without a scraping infrastructure. Ethical scraping that respects robots.txt means data is sourced within the rules the web has set — so you can sleep better at night. Clean text output reduces preprocessing overhead before tokenization.
Track competitor pages for changes.
Schedule periodic fetches of competitor pricing pages, documentation, or news sources. Structured output makes diff-ing and alerting straightforward — call on a schedule, pay only when a fetch fires, and act when something changes.
Stop building scrapers from scratch.
Connect a wallet, pass a URL, and your first page fetch lands in seconds. If your agent doesn't crawl for a month, you don't pay for the month. No infrastructure to maintain, no proxies to rotate, no robots.txt logic to write. That's the entire deal.
- 100% robots.txt compliance
- No scraping infrastructure
- MCP + pay-per-call native
- Per-call billing
The honest answers.
If something below doesn't cover your case, ping us — we answer directly, no SDR funnel.
What data does the endpoint return?
+
Three things in one response: clean readable text (suitable for embedding or LLM context), the full HTML source, and structured metadata — title, meta description, canonical URL, Open Graph tags, and heading hierarchy (H1–H3).
Does it respect robots.txt?
+
Yes, always. Every fetch checks robots.txt before retrieving the page. If the page disallows crawling, the endpoint returns a clear error rather than bypassing the directive. Ethical by default, not as an option.
How is billing calculated?
+
Per fetch — one charge when a URL is requested, regardless of page size or the number of output formats returned. If your agent doesn't crawl for a month, you don't pay for the month.
Can autonomous AI agents use this?
+
Yes — that's the primary design point. MCP discovery and pay-per-call payments let an autonomous agent find the endpoint, fetch a URL, and consume the response without a human setting up credentials or managing a scraping account.
Does it handle JavaScript-rendered pages?
+
Yes. The gateway renders JavaScript before returning the page content, so single-page applications and dynamically loaded content are captured in the same call.
Are there rate limits?
+
Per-key rate limits apply with concurrency-fair queueing — a high-volume agent never starves another agent on the same key. Practical limits are well above what individual agents typically need.