Runo is a scraping API that returns schema-defined typed JSON directly, with auto JS rendering, bot bypass, and batch/crawl modes built in.
Key Takeaways
Schema is plain JSON (field name, type, example, optional hint); no CSS selectors or XPath required.
Three endpoints: /extract (single URL), /batch (fan-out), /crawl (recursive, quota-capped at 25% of monthly balance per job with unused-page refunds).
Auto-escalates from plain HTTP to Playwright headless on JS-gate or 403; render_mode field in response shows which path ran.
Flat per-request pricing vs. credit-multiplier models; Scale tier quoted at $0.90 per 1K requests vs. ~$6 for Firecrawl equivalent.
Unresolvable fields return explicit null; type coercion enforced across string, float, integer, boolean, date, and array types.
Hacker News Comment Review
Commenters with production scraping experience want a live demo endpoint to test real URLs, see actual latency, and confirm bot detection tier before committing.
Open questions focus on gaps in the current API: no mention of cookie injection or UI interaction simulation (clicks, form fills), which matter for authenticated or multi-step scraping.
Hardware-attested CAPTCHAs (Google next-gen reCAPTCHA) were flagged as a likely ceiling for the bypass claims; no response from the builder yet.
Notable Comments
@drewrbaker: Asks whether cookie supply and UI interaction simulation are supported, two common requirements for authenticated scraping workflows.
@rvz: Raises hardware-attestation reCAPTCHA as a potential hard limit for the bot bypass tier.
