Site Audit

Crawl your site, run 156 technical SEO checks, see what changed since the last crawl, and fix issues with AI or a coding agent.

Site Audit crawls the active project's website on your own machine and checks every URL it finds. Open it from SEO → Site Audit in the sidebar. Since 1.6.0 a crawl can cover up to 10,000 pages and runs 156 checks across indexability, links, redirects, content, sitemaps, robots.txt, hreflang, social tags, duplicates, structured data, Core Web Vitals and AI discoverability.

The view has seven tabs: Overview, Issues, Pages, Links, Structure, Crawl and History.

Run a Crawl

Open the Crawl tab with the right project selected.

  1. Check the Start URL. It defaults to the project's website.
  2. Pick Max pages (25 up to 10,000) and a Speed (30 to 600 URLs per minute, or Unthrottled for your own staging server).
  3. Choose what runs After the crawl (see below).
  4. Click Run crawl and watch progress in the crawl log under the form.

The crawl reads robots.txt and your sitemaps, follows internal links, and records every URL, link, redirect, image, script and stylesheet it meets. The switches under the form control that: Respect robots.txt, Seed from sitemap, Follow nofollow links, Remove URL parameters, Check images, CSS & JS (up to 10 per page) and Check external links.

Advanced Options

Open Advanced for:

  • Scope: Exact host, Host + subdomains (www and the bare domain count as one site), or This path only.
  • Max depth and User agent (Mobile or Desktop).
  • Also start from these URLs (one per line): pages your links do not reach, such as campaign landing pages.
  • Also start from the pages Search Console shows in search: finds orphan pages and old URLs Google still knows. Needs the project's Search Console property.
  • Password-protected site (HTTP basic auth): save a username and password once per project to crawl a staging site. The password stays in your system keychain and is only sent to that site, over HTTPS.

Turning off Respect robots.txt only works for a site you can prove you own, through its Search Console property or your IndexNow key file. Otherwise robots.txt is always respected, and the form tells you why.

Optional Checks After the Crawl

  • Render JavaScript loads pages in a sandboxed browser and flags content and links that only appear after scripts run.
  • Core Web Vitals fetches real-user LCP, INP and CLS from PageSpeed Insights for your most-linked pages. It uses your PageSpeed key when one is saved.
  • Search Console traffic joins 28 days of clicks, impressions and position to flag broken, redirected or noindexed pages that still get traffic, traffic drops, and pages that fell out of the top 10.

Resume a Crawl

If a crawl stops early because you closed the app, clicked Cancel or hit a time limit, its page shows Resume crawl. It continues from where it ended instead of starting over.

Read the Results

Overview shows health, the share of crawled URLs without errors, so it moves as soon as you fix things. It also shows a health trend across crawls, the top issues, status codes by crawl depth, and how many sitemaps and URLs were read. Use Compare with to pick the earlier crawl to compare against.

Issues lists every issue found, grouped as errors, warnings and notices. For each issue you see how many URLs have it, a trend, and what changed since the compared crawl:

  • Added: existing URLs that now have the issue.
  • New: URLs new to this crawl that have the issue.
  • Removed: URLs where the issue is fixed.
  • Missing: URLs that had the issue but were not found this time.

Open an issue to see Why it matters, How to fix, and every affected URL. When one link or template is behind hundreds of affected pages, such as a dead footer link or a docs subdomain that no longer exists, the issue names that shared root cause.

Explore Pages, Links and Folders

  • Pages is a table of every URL with a filter builder and a column picker. Combine conditions, then click Save as custom issue to have every later crawl check that filter as its own issue.
  • Links lists every link with its kind, such as hyperlink, image, JavaScript, canonical, hreflang, redirect or iframe, and the top anchor texts for a target.
  • Structure groups URLs by folder with pages, indexable pages, errors, average depth and average load time per folder.

Click any URL to open its detail panel: status, indexability, redirect chain, structured data and rich-result validation, the issues on that URL, and its history across earlier crawls.

Export CSV on the Issues, Pages and Links tabs saves exactly what you see. Exports need a Pro license.

Fix Issues

From an issue's page:

  • Copy fix brief copies a Markdown brief with the issue, why it matters and the affected URLs.
  • Ask AI CMO opens the assistant with a fix plan for the issue.
  • Copy agent prompt gives a coding agent working in your site's repository a prompt that reads the issue through the 1MarketingTool MCP tools.
  • Request indexing asks Google to crawl the indexable pages in the issue (up to 100) with your Search Console service account. See Request Indexing.

AI agents connected through AI Access (MCP) can use the Site audit tool group to list audits, read issues and URLs, get fix briefs and start crawls.

Tune the Checks

The Issue settings section on the Crawl tab tunes the checks for this project. Turn a check off, change its level (error, warning or notice), or adjust limits like title length and slow-page time. You can also change a check's level from its issue page with Level for this project. Saving re-checks the latest crawl right away; checks you turn off are skipped and do not count toward health.

Schedule Crawls and Alerts

Under Schedule & automation on the Crawl tab, set Repeat to daily, weekly or monthly and pick a time. Scheduled crawls use the settings above. Two options act on what changed:

  • Alert me when health drops or new errors appear sends a notification.
  • Auto-submit new and changed pages to IndexNow submits them after each crawl. See IndexNow.

Click Save schedule to keep it.

Storage and Backups

The last five crawls of each site stay complete. Older crawls keep their summary, issue counts and trend, so the History tab and health trend still work. Change how many crawls stay complete under Settings → Storage (Keep site audits in full). Backups carry site audits as summaries unless you turn on Include full site audit detail when you export.

When to Re-run

Run a crawl after fixing issues, shipping a redesign, changing URL structure, or publishing a large batch of new pages. A weekly schedule catches regressions without you remembering to check.