Website audit
Technical and on-page SEO audit of one website, page by page, in a real browser.
Each page comes back with a 0 to 100 score and nine category scores, plus meta tags (title and description with lengths, canonical, robots, Open Graph, Twitter card, viewport, language), heading structure, content signals (word count, readability, keyword density, thin content, search intent), technical checks (HTTPS and certificate, mobile friendliness, sitemap, robots.txt, JSON-LD, microdata and RDFa types, AMP, indexability, whether the page needs JavaScript to render its content, pagination), page timing measured in the browser (time to first byte, first contentful paint, total blocking time, cumulative layout shift, speed index), links (internal and external counts, anchor text quality, broken links and redirect targets with the status code each returned), image alt and size coverage, and basic accessibility (ARIA labels, lang attribute, skip links, contrast). Every issue is classified critical, warning or info and carries an id, a category, a priority, an estimated impact and a fix hint. A final item of type site-summary averages the pages, ranks the issues that repeat across them, and reports crawl coverage.
What a call costs
Billed on actual usage, minimum $0.0020 per call, capped at $8.00 per call.
a price before every call. failed calls are never charged.
When to use it
Use when you have a website and need to know what is actually wrong with it: auditing a prospect before pitching, checking a site after a launch or migration, or turning a domain into a concrete fix list. One site per call, so pass a single URL and call again for the next one. Billed per PAGE audited, not per site: leaving maxPages out audits up to 5 pages, and whatever you set is what you pay for, up to 25. It does NOT tell you what a site is built on (no CMS, framework, analytics or hosting detection) and returns nothing about backlinks, keyword rankings or traffic.
Inputs
Website audit takes 16 fields, 1 of them required. Anything you leave out falls back to the tool's own default.
| Field | Type | Required | What it does |
|---|---|---|---|
| startUrls | array | yes | The one site to audit, as a URL string or an object with a url. One site per call: the page cap below is applied per start URL, so a second URL would double what you are billed while the quote still sized itself on one. Call again for the next site. |
| maxPages | integer | optional | How many pages to crawl and audit, and therefore how many pages you are billed for. The actor audits 5 when you leave it out. A site with fewer pages than this bills only the pages it found. 25 is Goro's ceiling, not the actor's: past it a crawl can outlive this endpoint's 20 minute run limit. |
| auditLinks | boolean | optional | Internal and external link counts, anchor text quality, broken links and redirect targets with status codes. On unless you set it false. |
| crawlPages | boolean | optional | Follow internal links from the start URL and audit more pages of the same site. The actor does this when you leave it out. Set it false to audit only the exact URL you passed, which bills one page. |
| auditImages | boolean | optional | Alt attributes, dimensions, file size, format, lazy loading. On unless you set it false. |
| auditSchema | boolean | optional | JSON-LD, microdata and RDFa detection with the schema.org types found. On unless you set it false. |
| auditContent | boolean | optional | Word count, readability, keyword density, thin content, search intent. On unless you set it false. |
| auditHeadings | boolean | optional | Heading hierarchy H1 to H6, multiple H1s, skipped levels. On unless you set it false. |
| auditMetaTags | boolean | optional | Title, description, Open Graph, Twitter card, canonical. On unless you set it false. Turning a section off makes the output smaller, never cheaper: the price is per page audited whatever it contains. |
| auditTechnical | boolean | optional | HTTPS and certificate, canonical, robots meta, sitemap, robots.txt, mobile friendliness, indexability, JavaScript rendering, AMP, pagination. On unless you set it false. |
| maxConcurrency | integer | optional | Pages audited in parallel. The actor uses 5. Lower it on a site that rate-limits you. |
| auditPerformance | boolean | optional | Page timing measured in the browser: time to first byte, first contentful paint, total blocking time, cumulative layout shift, speed index. On unless you set it false. Note it does not measure LCP or INP. |
| respectRobotsTxt | boolean | optional | Skip URLs the site's robots.txt disallows. The actor does this when you leave it out. |
| includeSubdomains | boolean | optional | Also crawl subdomains of the start URL, e.g. blog.site.com when starting from site.com. Off unless you set it. |
| auditAccessibility | boolean | optional | Basic checks only: ARIA label count, lang attribute, skip links, a contrast pass. Not a WCAG conformance audit. On unless you set it false. |
| excludeUrlPatterns | array of string | optional | URLs to keep out of the crawl. Plain substrings by default ("/tag/" drops anything containing it); wrap in slashes for a regex ("/\\.(pdf|zip)$/"). Worth setting on a CMS, where /tag/, /category/ and /author/ pages will otherwise eat the page budget you are paying for. |
How to call it
Connect once and your agent can reach Website audit the same way it reaches everything else in the catalog. Add Goro as an MCP connector, or paste the setup line into an agent that reads skills.
MCP
Add this server in Claude, Claude Code, or any MCP client.
https://mcp.usegoro.ai/mcp
Skill
One line in your agent's chat. It sets itself up.
set up https://usegoro.ai/SKILL.mdOr plain HTTP
POST /v1/run
{
"endpoint": "website.audit",
"input": {
"startUrls": [
"<startUrls>"
]
}
}Related tools
More from web and site audits, or browse the full tool catalog.
Further reading: Web scraping for AI agents: the practical guide
Run Website audit from your agent.
One balance for every tool in the catalog. Start with Build, $25 of credit for $19 a month.
Get started