Index Doctor: why your pages are not indexed, and the cheapest fix

Ask it in your own words

• "Why isn't Google indexing my pages?"
• "Half my site is "Discovered - currently not indexed". Is that bad?"
• "How do I get indexed faster? Is IndexNow worth it?"
• "my new site is not on google and i submitted the sitemap 2 weeks ago"

What happens next

1. It measures first. The ratio of indexed to submitted pages, dated. Pages too recent on the day of the reading are marked "too early to judge", not diagnosed.
2. It sorts statuses, then tests. Each excluded URL goes to a cause class. triage.py, the triage script, then fetches your pages and records the literal evidence of each technical cause.
3. It ranks fixes by template. At most five groups make the plan. Acceleration follows the fixes, never replaces them.

What you get back

• A three-line verdict: main cause class, how sure, first action
• The indexation ratio (date, source, trend) and a breakdown by status
• A fix plan by group: cause, evidence, fix, effort, confidence, checkpoint
• Safe acceleration steps and a plan to measure again

Run on this very site

A triage of the 35 URLs of the sitemap, with no Search Console export to read yet.

Indexation ratio: Not measurable here. [...] comes from Search Console's Page indexing report
Age gate (heuristic: URLs younger than 21 days at the time of the reading are not judged [...])
WARNING: 35 of 35 sitemap lastmod values are the same date (2026-10-12) [...] That date is not a page age and was not used as one.
noindex served (meta or header): 0 URLs [...] Sitemap hygiene: 0 listed URLs are not clean indexable 200 pages.

On a domain two days old there is nothing to conclude about indexing, and the skill said so instead of guessing. It will be run again on the first Page indexing export.

What it will not do

• It cannot say whether Google will index a page, or why it left one out: Search Console gives labels, not reasons.
• Without Search Console, the ratio and the Google-side statuses cannot be measured. The crawl still shows what is wrong on the pages.
• IndexNow reaches the engines that read it, such as Bing and Yandex. Google was not among them when the skill was last checked (10 October 2026).

What it needs

Claude Code, Claude.ai, Claude Desktop or Cowork. The two scripts need code execution with Python 3.9+ and web access to your site; without them, every check has a manual version, run on a sample.

How to install a skill in Claude

Guides

• How to use Claude for SEO: which Claude for which job, and what data to give it
• Lots of impressions, almost no clicks in Search Console: what to check, in order
• Discovered - currently not indexed: what it means and what to do, in order

Where it hands over

• Link Weaver: When the pages left out are orphans, the fix is a link design.
• Street Cred: On a young domain, a few real third-party mentions do more for crawl demand than more pages.

Questions about this skill