How to use the AI Search GEO Optimization agent skill (seo-geo)

What is an agent skill, and what does seo-geo make the agent do?

An agent skill is a folder containing a SKILL.md file of instructions that an AI agent loads when a request matches the skill's description. The seo-geo skill is named seo-geo, is licensed MIT, and is published by AgriciDaniel in the claude-seo repository.

Once installed, the agent evaluates a page or site against criteria including self-contained answer blocks, heading structure, author and date signals, brand mentions, and whether AI crawlers can access the page according to robots.txt. It then reports findings framed as ordinary SEO fundamentals and notes where community advice conflicts with Google's own guidance.

• Citability: whether passages can be quoted without extra context (weighted 25% in the skill's criteria)
• Structural readability: heading hierarchy, question-based headings, short paragraphs, tables and lists (20%)
• Multi-modal content: images, video, charts and interactive elements (15%)
• Authority and brand signals: bylines, dates, primary-source citations and presence on platforms such as Reddit, YouTube and LinkedIn (20%)
• Technical accessibility: server-side rendering, AI crawler access in robots.txt and RSL 1.0 licensing terms (20%)

How do you install the seo-geo skill?

Download the ZIP for the skill from the download link on this page. Then choose one of the two methods below depending on how you use your agent.

For Claude Code, unzip the file into your personal skills folder ~/.claude/skills/, or into a project's .claude/skills/ folder if you want the skill only for that project. The skill should sit in its own folder containing SKILL.md.

In the Claude apps, you can upload the ZIP instead under Settings → Capabilities → Skills.

• Personal install: ~/.claude/skills/
• Project install: .claude/skills/ inside your project
• Claude apps: Settings → Capabilities → Skills

How do you trigger the skill with a natural request?

You do not need a special command. The skill's description tells the agent to use it when you ask about AI Overviews, AI Mode, ChatGPT, Perplexity or Gemini citations, GEO, or AI search visibility. Describe what you want in plain language and include the page address if you have one. The skill's argument hint is [url].

Ask for a specific page when you want a page-level review, and for a site when you want a crawler and robots.txt check across the whole domain.

• "Audit https://example.com/guides/returns for AI Overviews citability."
• "Check whether my robots.txt blocks AI crawlers that affect ChatGPT Search and Claude search."
• "Review this article for quotable, well-sourced structure before I publish."
• "Is my pricing page likely to be cited by Perplexity? Give me the main fixes."

What should you expect in the report?

The report groups findings under the skill's criteria and frames them as SEO fundamentals applied to AI search surfaces, not as a separate discipline. Where community recommendations contradict Google's primary guidance, the skill defers to Google and says so in the report.

Read the scores as heuristics. The skill's own material states that the scores are not guarantees, and that statistics it cites are third-party figures that were not re-verified by the authors. Quote them only with their source and date, or leave them out.

• llms.txt is reported for completeness but carries no weight in the score, because Google Search ignores it
• Crawler findings separate training bots (such as GPTBot and ClaudeBot) from search and user-triggered bots (such as OAI-SearchBot and Claude-SearchBot)
• Google-Extended controls Gemini training and grounding, and does not control Google Search inclusion or ranking

What does the skill need, and what are its safety notes?

The skill lists no required prerequisites. It does not run scripts, but it does use the network, so the agent may fetch the page or robots.txt you ask it to check. Review the skill's files before installing if your environment has rules about outbound requests.

The skill does not account for a business's legal or licensing position. Do not block or allow crawlers based only on this report. Check with the people responsible for your content and legal terms first.

• Requires: nothing listed
• Runs scripts: no
• Uses network: yes
• Files included: SKILL.md, LICENSE.txt, references/google-ai-optimization-guide.md, references/llmstxt-evidence.md

What tips make the audit more useful?

Start with the fundamentals the skill reports on: give each important answer a self-contained opening, use question-based headings, attribute claims to specific sources, and show a clear byline and dates. The skill's material also says that pages left stale for 6+ months lose citation eligibility according to one third-party study, so a regular refresh schedule is worth considering.

Check robots.txt rules deliberately. Training crawlers and search crawlers are different bots, so blocking one does not necessarily affect the others.

• Put the most citable answer near the top of the page, since one third-party study found about 44% of citations come from the first 30% of a page
• Do not rewrite content to chase llms.txt; Google says it neither helps nor harms visibility in Google Search
• Re-run the audit after changes and compare the findings rather than the raw score