CAPTCHAs, solved inside the scrape
When a page throws a challenge, Spidra solves it in the same job and keeps reading. There's nothing to switch on, and you only pay for a solve that worked.
Twelve kinds of challenge, three families
Spidra recognizes the challenge type and sends it to the right solver. Supported means handled on most sites, not guaranteed on every one.
Checkbox and image puzzles
The ones a person would click through.
- reCAPTCHA v2 (checkbox)
- reCAPTCHA v2 (invisible)
- hCaptcha
- GeeTest
- FunCaptcha / Arkose Labs
- AWS WAF CAPTCHA
Invisible, score-based checks
No puzzle on screen; the page decides in the background.
- reCAPTCHA v3
- Cloudflare Turnstile
Bot-protection challenge pages
The interstitial pages that stand in front of a site.
- Cloudflare JS challenge
- DataDome
- PerimeterX / HUMAN
- Imperva / Incapsula
Nothing in the request. A count in the result.
The request is an ordinary scrape. The result tells you how many challenges were solved, so you can see what each page cost.
Request: no CAPTCHA settings
{"urls": [{ "url": "https://…" }],"prompt": "Extract the main content","output": "json"}
- Same on every tool. Scrape, batch scrape and crawl all solve challenges they meet.
- Pair it with Stealth Mode when a site keeps showing challenges, so the request comes from a residential connection. How Stealth Mode works
Response: the example from the docs
{"status": "completed","result": {"content": { … },"stats": {"durationMs": 12400,"captchaSolvedCount": 1,"inputTokens": 2800,"outputTokens": 140,"totalTokens": 2940}}}
captchaSolvedCount: 0 means there was no challenge, or the page let the browser through without one, which is common with Cloudflare's JavaScript check. Each page in result.data carries its own count too.
Some sites still win. You don't pay when they do.
No scraping service reads every site every time. When a challenge can't be cleared, the job fails with a plain reason, and neither the page nor the solve attempt is charged.
What a page costs
- No CAPTCHA1 credit
- One CAPTCHA solved6 credits
- Two solved (on load and after a click)11 credits
- Challenge still there after a solve0 credits
AI extraction, if you use a prompt or schema, is added the usual way.
Questions, answered
CAPTCHA solving
No. It's always on, on every scrape, batch scrape and crawl, with no parameter. If a page has no challenge, nothing happens and nothing extra is charged.
You build the product.
Spidra handles the data work.
Start with 300 free credits, about 300 pages. No credit card required.