Paste a URL
Any site, any stack. There's nothing to install first.
Reach / Reports▊
Full-site audits that tell you what to fix next.
Crawl the whole site, see what's getting in the way, and get a clear fix for every issue. No heavy desktop app. No afternoon lost in a spreadsheet.
Latest crawl · 8 minutes ago
acme.com
Health82+7 since last crawl
GEO74
Accessibility96
Fix these firstImpact
All 18Errors 4Warnings 9Notices 5
Broken internal links
These links lead visitors and crawlers to a dead end.
12 pages affected · estimated score lift +5PageStatusSEO
/pricing20094
/docs/getting-started20088
/old-features40442
Let Crawlie look around
The crawl runs on our servers, so your laptop stays quiet.
Read the fixes
Prioritised issues in plain English, shareable with one link.
Why it wins
Built different where it counts.
Three scores that can't hide each other
Health, GEO and Accessibility stay separate. A tidy sitemap can't hide pages that nobody can use or cite.
148 checks, each with a fix
See why an issue matters, how to fix it, and every page it touches. Then hand it to the person who owns the work.
Six-figure crawls, browser-light
Large crawls run in the cloud and open like any other report. Your browser gets the useful bits, not the burden.
GEO
Audit for AI search, not just Google
ChatGPT, Perplexity and AI Overviews read your site differently. Crawlie checks the signals that decide whether you get cited: answerable content, structured data, authorship, llms.txt, semantic markup.
- GEO score per page and site-wide
- llms.txt, robots and sitemap discovery
- 23 schema types validated against rich-result specs
Answer-ready content yes
Structured data Product · FAQPage
Authorship missing
llms.txt found
GEO score 75 / 100
Share & export
Reports that travel
Send a public link no login can gate, export issues as CSV or the whole report as JSON, and let your agents pull everything over the API and MCP.
- Public share links, revocable any time
- CSV / JSON export
- Full report over REST + MCP for agents
{
"healthScore": 62,
"geoScore": 75,
"topFixes": [
{ "rule": "broken-link", "count": 214 }
]
}
The whole report
One crawl. Eight ways into the answer.
Start with the big picture, then follow the problem all the way down to the page, redirect or link that caused it.
Guide: reading a report →Health82
GEO74
Accessibility96
Broken internal links12 pages · fix the template once
High impactPages missing authorship6 pages · GEO
MediumLong title tags4 pages · search snippets
LowCloud by default
An audit that doesn’t stop at your laptop.
The crawl keeps running when you close the tab. The finished report is ready for a teammate, a client or an agent without another export ritual.
Run a report free →Crawlie CloudDesktop crawler
Crawl after the tab closesIncludedKeep the machine awake
Share a live reportOne linkExport and upload
Compare every runAutomatic diffsManual project files
Give an agent the dataREST + MCPBuild the bridge
Deploy #28412 issues resolved
Fixed
✓ Missing canonical 38 pages
✓ Broken navigation link 12 pages
✓ Product schema price 8 pages
New
+ Long title tag 2 pages
+ Missing author 1 page
Before and after
Proof the fix actually landed.
Compare any two crawls for score changes, pages added or removed, and issues that appeared or disappeared. Older reports tell you when a newer one exists, too.
- Automatic comparison after every scheduled crawl
- Structured diffs through the API and MCP
- Public links you can revoke whenever you like
Agent-native
Your agents get the same superpowers.
Everything on this page is a tool call away. The hosted MCP server exposes 18 tools — crawls, reports, fixes, monitoring — and the REST API speaks the same JSON. One key, no scraping, no glue code.
- Hosted MCP endpoint for Claude, Cursor and any MCP client
- REST API with the same capabilities, self-describing
- Open-source CLI + local MCP if you'd rather stay on your machine
> audit acme.com and fix the top issues
crawl_site { "url": "acme.com" }
✓ report ready · health 62 · geo 75
top_fixes → 3 fixes ranked by impact
affected_urls broken-link → 214 pages
explain_issue why it matters + the fix
18 tools · REST API · OAuth or API key
What does a Crawlie SEO report check?
Every crawl runs 148 checks across 17 categories: technical SEO, indexability, links, redirects, titles and meta, canonicals, structured data, performance, security, accessibility (WCAG) and GEO — the signals AI search engines read. Each issue comes with a plain-English explanation of why it matters and how to fix it.
Can Crawlie audit JavaScript-heavy sites like React or Next.js?
Yes. Turn on JavaScript rendering and Crawlie drives a real headless Chrome, lets each page hydrate, and audits the post-JavaScript DOM — so content, links and meta tags injected by React, Next.js or Vue are seen instead of missed.
What is a GEO score?
GEO (Generative Engine Optimization) measures how ready your pages are to be cited by AI search — ChatGPT, Perplexity and Google AI Overviews. Crawlie checks answerable content, structured data, authorship, llms.txt and semantic markup, and scores each page and the whole site.
How large a site can Crawlie crawl?
Crawls run in the cloud, so six-figure page counts open like any other report. The free Hobby plan covers 1,000 pages per crawl; Pro raises that to 50,000 pages per crawl, with monthly credits to match and prepaid top-ups beyond.
Can I share a report with my team or a client?
Every report has a public share link that needs no login and can be revoked any time. You can also export issues as CSV, the whole report as JSON, or let agents pull everything over the REST API and MCP.
Point Crawlie at your site.
Free to start on the Hobby plan — 5,000 crawl credits a month, no card required.