Give your AI agent eyes on the web

hintrix lets AI agents read any website, extract structured data, crawl entire domains, and audit pages for AI search visibility. One API call returns clean Markdown and GEO diagnostics with actionable fixes.

request & response
$ curl -X POST https://hintrix.com/v1/scrape \
  -H "X-API-Key: hx_live_sk_..." \
  -d '{"url": "https://example.com", "mode": ["content", "audit"]}'

{
  "agent": "reveal",
  "content": {
    "markdown": "# About Us\n\nWe build tools...",
    "word_count": 890
  },
  "audit": {
    "geo_score": 72,
    "tech_score": 85,
    "issues": [...]
  },
  "credits_used": 2
}
// npm install hintrix
import { Hintrix } from 'hintrix';

const hx = new Hintrix('hx_live_sk_...');
const page = await hx.scrape('https://example.com', {
  mode: ['content', 'audit']
});

console.log(page.content.markdown);
console.log(page.audit.geo_score); // 72

// Crawl an entire site and collect all pages
const { pages } = await hx.crawlAndCollect('https://example.com', {
  max_pages: 50,
  onProgress: (s) => console.log(`${s.pages_crawled} pages...`),
});
Content

LLM-ready output from any page

Clean Markdown, HTML, or plain text. Metadata, links, structured data, and Schema.org included.

  • Static HTML and JS-rendered pages
  • JSON APIs and data endpoints
  • SPAs, React, Next.js, Vue
  • Automatic JS detection
GEO Audit

Will AI search engines cite this page?

GEO readiness score with 80+ evidence-backed checks. Issues ranked by impact with copy-paste fixes.

  • AI bot accessibility analysis
  • Citation readiness scoring
  • E-E-A-T and entity signals
  • Structured data validation
01

Read any website as context

Your agent scrapes a URL and gets clean Markdown that fits directly into a prompt. No parsing, no HTML cleanup.

02

Research across entire domains

Crawl a full website and collect content from every page. Build knowledge bases, feed RAG pipelines, or summarize documentation.

03

Extract structured data

Pull prices, products, contacts, or any data as JSON from any page — including SPAs, JSON endpoints, and JS-rendered content.

04

Audit AI search visibility

Check if a page will be cited by Perplexity, ChatGPT Search, or Google AI Overviews. Get a score, issues, and copy-paste fixes.

05

Handle JavaScript-heavy pages

SPAs, React apps, and client-rendered pages that return empty HTML to normal crawlers. Rendered automatically when needed.

API endpoints

Five endpoints. Content extraction, auditing, structured data, batch processing, and multi-page crawls.

POST
/v1/scrapeagent: glance | reveal
Single URL to Markdown, HTML, or text. Optionally include GEO audit. Handles JS rendering automatically when needed.
1 credit (content) · 2 credits (content + audit) · +1 screenshot
POST
/v1/auditagent: reveal
GEO readiness audit with scores, issues, and fixes. Checks AI bot access, citation readiness, E-E-A-T signals, structured data.
2 credits
POST
/v1/extractagent: pinch
Structured data from any page. Define a schema with CSS selectors or let auto-detection handle it. Works with SPAs and JSON endpoints.
2 credits
POST
/v1/batchagent: sweep
Submit multiple URLs in a single request. Async job processes them in parallel with a single status endpoint to poll.
same as per-URL cost · billed per item · +1 screenshot per URL
POST
/v1/crawlagent: sweep
Multi-page crawl with depth control. Async job with progress tracking. Content and audit for entire sites.
1 credit / page · +1 audit · JS rendering included

Capabilities

What you can do with hintrix.

Markdown outputClean, structured content ready for any LLM pipeline
GEO scoringEvidence-backed readiness score for AI search visibility
JS renderingFull browser rendering for SPAs, detected automatically
JSON & API contentRead and parse JSON endpoints, RSS feeds, sitemaps
Schema.org extractionJSON-LD, microdata, and RDFa parsed and returned
Fix suggestionsCopy-paste fixes with severity, impact, and effort ratings
Node.js SDKnpm install hintrix — typed client with retries and polling helpers
MCP serverNative integration for Claude Code, Cursor, and Windsurf
robots.txt analysisWhich AI bots are allowed, blocked, or rate-limited
SSRF protectionMulti-layer defense against server-side request forgery
PageSpeed scoresCore Web Vitals and performance metrics alongside audit results
ScreenshotsOpt-in full-page screenshots for visual verification and diffing (+1 credit, not stored)
Content diffingCompare page content across scrapes — see what changed. Previous versions stored 7 days per URL. No extra cost.
Link health reportsBroken link detection across pages and entire domains
llms.txt generationAuto-generate and audit llms.txt files for AI-agent discoverability

Pricing

Pay per use. No subscriptions. Credits never expire.

Free

$0
500 credits
on signup, no card

$5 pack

$5
2,500 credits
$0.002 / credit

$12 pack

$12
7,500 credits
$0.0016 / credit

$29 pack

$29
20,000 credits
$0.00145 / credit

Credit costs

ActionCredits
/v1/scrape1
/v1/scrape + audit2
/v1/audit2
/v1/extract2
/v1/batchper-URL cost
/v1/crawl1 / page  (+1 audit per page)
JS renderingincluded

Cost examples

Scrape 1,000 blog posts for a RAG pipeline1,000 cr · $2.00
GEO audit 100 pages for a client200 cr · $0.40
Crawl a JS-rendered shop, 500 pages500 cr · $1.00
Extract product data from 200 listings400 cr · $0.80

What makes hintrix different

Most crawl APIs give you content or diagnostics. hintrix returns both in a single request.

Content + diagnostics, not either/or

Other tools return raw Markdown or SEO data. hintrix gives you clean, LLM-ready content and a GEO audit with scores and fixes from the same API call.

Pay per use, not per month

No subscriptions. Buy credit packs when you need them. Start with 500 free credits on signup, no credit card required — get 500 more by sharing on X. Credits never expire.

Built for AI agents

MCP server for Claude Code, Cursor, and Windsurf. Markdown output optimized for LLM context windows. Auto-detection of JS rendering needs.

Actionable, not just data

Every issue comes with a severity rating, impact estimate, and a fix you can copy and paste. No interpretation needed.

Frequently asked questions

Common questions about hintrix.

What domains are blocked?

Social media platforms (Facebook, Instagram, Twitter/X, LinkedIn, TikTok, YouTube, Reddit, Pinterest, Snapchat, Threads) and dark web domains are blocked for legal and ethical reasons.

Do you respect robots.txt?

Yes, by default. You can override this per request with respect_robots_txt: false for URLs where you have explicit permission.

What User-Agent does hintrix use?

HintrixBot/1.0 (+https://hintrix.com/bot). Website owners can allow or block this in their robots.txt.

Do I need to verify my email?

Yes. A verification email is sent on signup. You must verify before making API calls.

What happens when JS rendering is needed?

hintrix uses full browser rendering by default for reliable content extraction. You can disable it with wait_for_js: false for faster plain HTTP scraping. No extra cost either way.

Do credits expire?

No. Credits never expire. Buy when you need them.

Can I get a refund?

All credit purchases are final and non-refundable.

Start with 500 free credits

No credit card required.

Create account Read the docs