seo-geo: SEO and GEO audit for Yandex, Google and AI assistants

seo-geo-skill

A Claude Code skill with a script and checklist: Yandex and Google indexing, Webmaster, Metrika and visibility in AI assistants

  • 2 ★ on the repository
  • Last commit
  • MIT
Skill

Minimal access

Why this level

  • The script crawls the chosen site with GET and HEAD requests and checks up to hundreds of links, which adds load on a third-party site
All reasons and checks
Russian stack

cryptarl/seo-geo-skill

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/seo-geo-skill

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Minimal accessThe tool runs nothing and makes no network calls.

  • The script crawls the chosen site with GET and HEAD requests and checks up to hundreds of links, which adds load on a third-party site

This is third-party code. Review the repository files before installing.

What it does

The seo-geo skill checks a site in a few minutes and writes a SEO_AUDIT.md report with P0 to P2 priorities. The seo_check.py script, plain standard-library Python, reads the site like a bot: it checks whether Googlebot and YandexBot get the page, HTTPS and mirrors, robots.txt (Clean-param for Yandex, blocked AI bots), sitemap.xml, llms.txt, and on a page sample noindex, canonical, title and description, H1, schema.org and Open Graph. Claude handles the rest from the project code and a manual eight-section checklist: Webmaster, Metrika and goals, Yandex Business, IndexNow, Dzen RSS, domain migration, answers in ChatGPT, Alice and GigaChat.

Who it is for. Owners and developers of sites on Next.js, Lovable, v0 and Bolt who need to know why Yandex and AI assistants do not see their pages.

Good fit when

  • Before launch a site has to be prepared for indexing in Yandex and Google
  • The site was built in Cursor or Lovable and a localhost address ended up in canonical
  • Yandex does not see pages behind Cloudflare or a captcha and the cause is unclear
  • You want the brand in Alice, GigaChat and ChatGPT answers and need a plan

Not a fit when

  • You need query position tracking: the skill checks the technical part and the checklist, not rankings
  • A paid backlink audit is needed: the script does not analyze backlinks
  • The site is fully client-rendered and behind a login: the script reads HTML without JavaScript

Example request

Check the SEO of https://my-site.ru and explain why Yandex does not see part of the pages

Limitations

The script looks at a page sample and HTML without running JavaScript, so findings need rechecking: for example it may treat og:image served through Next opengraph-image as missing. The skill does not fetch Alice, GigaChat or ChatGPT answers itself, since they need a login or captcha, and the owner has to send screenshots. It does not read Webmaster or Metrika accounts through an API, so those items stay manual. There is one author, two stars and the last commit in September 2026. The skill makes no changes without permission.

How to disable. Delete the ~/.claude/skills/seo-geo folder and, if you like, the SEO_AUDIT.md and SEO_AUDIT.auto.md files in projects where the skill wrote reports.

FAQ

What does it need to run?

Claude Code and Python 3.9 or newer: seo_check.py uses only the standard library and installs nothing.

Does the skill change my site?

No, the script only reads the site with GET and HEAD requests, and Claude proposes fixes after the report and applies them only with permission.

Can I run the script without Claude?

Yes: python3 ~/.claude/skills/seo-geo/scripts/seo_check.py https://my-site.ru --report SEO_AUDIT.auto.md; --pages and --links set the scope and --json gives machine output.

Which search engines does it cover?

Yandex and Google, plus Bing as the source for ChatGPT; Webmaster, Metrika, Yandex Business, IndexNow and Dzen are covered separately.

What comes out?

A SEO_AUDIT.md file: a short summary, then P0, P1 and P2 tasks and a list of manual steps for the owner in the consoles.

Discussion

No questions or reviews yet.

Yandex SEO audit

yandex-seo-audit

Russian stack

A Claude Code skill for technical SEO audits under Yandex requirements: canonical tags, JSON-LD, robots.txt, favicon and Webmaster diagnostics

SkillExtended access1Repository stars
Russian stack

A Claude Code plugin with 15 skills and 10 parallel subagents: a full SEO audit including checks for Yandex Turbo Pages, Metrika and robots.txt

PluginExtended access4Repository stars
Russian stack

A local SEO audit skill for the CIS market: business cards in Google Business Profile, Yandex Business, Yandex Maps, 2GIS and their consistency

SkillExtended access3Repository stars
Russian stack

A Claude Code skill for Yandex Metrica built on bash scripts over curl: preset reports, goals, counters and raw log exports

SkillExtended access1Repository stars
Russian stack

Five cache-first shell skills for Direct, Metrika, Webmaster, Wordstat and Search API, with Claude choosing which one to run

SkillBroad access17Repository stars

README in short

The README covers the script automatic checks, the part Claude does from code and the checklist, the Yandex and AI assistant sections, installation via install.sh, trigger phrases, running the script alone with parameters and a report fragment.

README badge

Report a problem