Technical SEO Audit: the Claude skill that finds what blocks your rankings
Technical SEO Audit is a free Claude skill that finds the technical issues stopping Google from reading your site: conflicting canonicals, redirect hops, indexing problems, duplicate titles and dead subdomains. Point it at a domain or a crawl export and it returns a prioritised fix list.
Install it
One command, then hand it a domain.
curl -o ~/.claude/skills/technical-seo-audit.md https://raw.githubusercontent.com/lhitches/claude-seo-skills/main/skills/technical-seo-audit.md
Then: "Run a technical SEO audit on example.com." No install? Paste the skill file contents into any Claude chat and the same process runs.
What it checks, and what you get
The audit an agency runs in week one of an engagement, structured by what to fix first.
- Canonical integrity: self-referencing canonicals, conflicts between canonicals and sitemap entries, variants leaking into the index.
- Redirects and status codes: internal links routed through redirect hops, dead subdomains, sitemap entries that redirect.
- Indexing intent vs reality: whether pages meant to be noindexed actually are, and whether indexable pages are in the sitemap.
- On-page hygiene at scale: multiple H1s, title and description lengths, duplicates, thin main content on key pages.
- The deliverable: Critical this week, high-priority this month, medium on the roadmap, five quick wins, a what-is-working list, and an honest not-checked section naming what needs a full crawler.
Real sample output
This excerpt comes from a genuine run against hawkacademy.co, my own education site. It found two critical issues we are fixing this week. Auditing your own property and publishing the result is the fastest way to show you the skill does not soften its findings.
TECHNICAL SEO AUDIT: hawkacademy.co
Data source: live fetches (robots.txt, sitemap.xml, 8 full pages, status checks on 17 URLs)
- CRITICAL 1: the main commercial page canonicals to a test variant. /workshop (sitemap priority 0.9) carries a canonical pointing at /workshop-j, a live indexable variant that self-canonicals. The sitemap tells Google to index one URL while the page nominates another, so Google chooses, and it may pick the variant.
- CRITICAL 2: the www subdomain is dead. www.hawkacademy.co returns a 522 with no redirect to the apex domain. Every www visit errors and any link equity pointed at www is lost. Fix: a 301 redirect rule, five minutes of work.
- HIGH: every internal link routes through a redirect. All sampled pages link to .html paths that 308-redirect to extensionless URLs (verified per hop). The homepage nav has 15 such links; one template fix removes a sitewide crawl tax.
- Quick win 3: /quiz has no robots meta despite robots.txt comments saying it should be noindexed. One line brings behaviour in line with stated intent.
Excerpt from a live run, 9 July 2026. Full report format: critical / high / medium priorities, top 5 quick wins, what's working, and an explicit not-checked list for full-crawl items.
When to use it, and when to use a sibling skill
Use Technical SEO Audit when you want the week-one agency audit on a live site.
| You want | Use | Why |
|---|---|---|
| A prioritised technical audit of any live site | Technical SEO Audit (this skill) | Works from live fetches, no crawler required, honest about limits |
| A repo-level audit with fixes staged as commits | Claude Code SEO | Audits the codebase itself and writes SEO_REPORT.md in git |
| A fast single-URL score before going deeper | Website Score | The opening move on every StudioHawk audit |
| Depth from a full crawl you already ran | Screaming Frog Analyser | Drop the export in, get the agency-style analysis back |
FAQ
Do I need Screaming Frog or a crawler to use this skill?
No. Give it a domain and it works from live fetches: robots.txt, the sitemap, key pages and status checks. Anything that genuinely needs a full crawl, like orphan pages or sitewide redirect chains, gets listed in a not-checked section instead of guessed at. For crawl exports, pair it with the Screaming Frog Analyser skill.
What do I give the skill to start?
A domain is enough. It fetches robots.txt, sitemap.xml and a sample of pages, then runs status checks on the URLs it finds. If you have a crawl export or Search Console data, drop that in too and the audit gets deeper on the same run.
How is this different from a paid auditing tool?
Paid tools list everything and leave the judgement to you. This skill runs the prioritisation an agency runs: critical this week, high this month, medium on the roadmap, plus five quick wins and a what-is-working list so you do not fix things that are fine. No subscription, no seat licence.
Will it catch everything wrong with my site?
No, and it says so. Every report ends with an explicit not-checked list naming what needs a full crawler: orphan pages, click depth, sitewide duplicate detection. An audit that admits its limits is one you can act on with confidence. The sample on this page shows the format.
Want the theory behind it? The Hawk Academy deep dive covers how StudioHawk structures technical audits.
Created by Lawrence Hitches
Chief of Staff at StudioHawk. I built this library from the processes our 120-person team runs across 500+ client campaigns. Every skill is free and MIT licensed.