Generated at build time from the skill catalog and each skill's NOTICE.md
GEO / AI Search SEO
- sourcehttps://github.com/zubair-trabzada/geo-seo-claude
- pinned commit989cae01e8eb
- licenseMIT
Audit and optimize websites for AI search citation (ChatGPT, Perplexity, Google AI Overviews): citability, AI crawlers, llms.txt, schema, technical SEO, reports.
Install it with agemon --skill-groups geo-seo, or pick it in the interactive prompt.
Skills
Section titled “Skills”| Skill | Description | Source | Commit | License |
|---|---|---|---|---|
geo |
GEO-first SEO analysis tool. Optimizes websites for AI-powered search engines (ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews) while maintaining traditional SEO foundations. Performs full GEO audits, citability scoring, AI crawler analysis, llms.txt generation, brand mention scanning, platform-specific optimization, schema markup, technical SEO, content quality (E-E-A-T), and client-ready GEO report generation. Use when user says “geo”, “seo”, “audit”, “AI search”, “AI visibility”, “optimize”, “citability”, “llms.txt”, “schema”, “brand mentions”, “GEO report”, or any URL for analysis. | https://github.com/zubair-trabzada/geo-seo-claude (geo/SKILL.md) | 989cae01e8eb |
MIT |
geo-audit |
Full website GEO+SEO audit with parallel subagent delegation. Orchestrates a comprehensive Generative Engine Optimization audit across AI citability, platform analysis, technical infrastructure, content quality, and schema markup. Produces a composite GEO Score (0-100) with prioritized action plan. | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-audit/SKILL.md (+ agents/geo-*.md as references/)) | 989cae01e8eb |
MIT |
geo-citability |
AI citability scoring and optimization. Analyzes web page content to determine how likely AI systems (ChatGPT, Claude, Perplexity, Gemini) are to cite or quote passages from the page. Provides a citability score (0-100) with specific rewrite suggestions. | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-citability/SKILL.md) | 989cae01e8eb |
MIT |
geo-crawlers |
AI crawler access analysis. Checks robots.txt, meta tags, and HTTP headers to determine which AI crawlers can access the site. Provides a complete access map and recommendations for maximizing AI visibility while maintaining appropriate control. | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-crawlers/SKILL.md) | 989cae01e8eb |
MIT |
geo-llmstxt |
Analyzes and generates llms.txt files – the emerging standard for helping AI systems understand website structure and content. Can validate existing llms.txt files or generate new ones from scratch by crawling the site. | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-llmstxt/SKILL.md) | 989cae01e8eb |
MIT |
geo-brand-mentions |
Brand mention and authority scanner for AI visibility. Analyzes brand presence across platforms that AI models rely on for entity recognition and citation decisions. Produces a Brand Authority Score (0-100) with platform-specific recommendations. | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-brand-mentions/SKILL.md) | 989cae01e8eb |
MIT |
geo-platform-optimizer |
Platform-specific AI search optimization — audit and optimize for Google AI Overviews, ChatGPT, Perplexity, Gemini, and Bing Copilot individually | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-platform-optimizer/SKILL.md) | 989cae01e8eb |
MIT |
geo-technical |
Technical SEO audit with GEO-specific checks — crawlability, indexability, security, performance, SSR, and AI crawler access | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-technical/SKILL.md) | 989cae01e8eb |
MIT |
geo-content |
Content quality and E-E-A-T assessment for AI citability — evaluate experience, expertise, authoritativeness, trustworthiness, and content structure | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-content/SKILL.md) | 989cae01e8eb |
MIT |
geo-report |
Generate a professional, client-facing GEO report combining all audit results into a single deliverable with scores, findings, and prioritized actions | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-report/SKILL.md) | 989cae01e8eb |
MIT |
geo-schema |
Schema.org structured data audit and generation optimized for AI discoverability — detect, validate, and generate JSON-LD markup | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-schema/SKILL.md) | 989cae01e8eb |
MIT |
geo-report-pdf |
Generate a professional PDF report from a GEO audit using pandoc + Chrome headless. Converts GEO-AUDIT-REPORT.md into a styled, client-ready PDF with a cover page, color-coded score tables, severity-tagged findings, and a 90-day roadmap. | https://github.com/zubair-trabzada/geo-seo-claude (skills/geo-report-pdf/SKILL.md (+ templates/)) | 989cae01e8eb |
MIT |
Subagents
Section titled “Subagents”| Agent | Tools | Purpose |
|---|---|---|
geo-audit-ai-visibility |
WebFetch |
Scores how quotable and visible a website is to AI search: citability, AI crawler access, llms.txt and brand presence. Used by the geo-audit skill; do not use it for other tasks. |
geo-audit-platform |
WebFetch |
Scores a website’s readiness for Google AI Overviews, ChatGPT, Perplexity, Gemini and Bing Copilot. Used by the geo-audit skill; do not use it for other tasks. |
geo-audit-technical |
WebFetch, Bash |
Scores a website’s technical health for AI crawlers: rendering, meta tags, headers, speed and security headers. Used by the geo-audit skill; do not use it for other tasks. |
geo-audit-content |
WebFetch |
Scores a website’s content quality and E-E-A-T signals: experience, expertise, authority and trust. Used by the geo-audit skill; do not use it for other tasks. |
geo-audit-schema |
WebFetch, Bash |
Scores a website’s schema.org structured data and finds missing GEO-relevant schema. Used by the geo-audit skill; do not use it for other tasks. |
NOTICE
Section titled “NOTICE”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: geo/SKILL.md- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.- Turned the orchestrator into a router: each `/geo <command>` names the skill to follow. Removed its own audit steps, the sub-skill directory table, the subagent table and the PDF section, which duplicated `geo-audit` and `geo-report-pdf`.- Removed the `/geo update`, `/geo prospect`, `/geo proposal` and `/geo compare` routes and their `~/.geo-prospects` output files (not vendored).- Replaced the market-statistics table with one neutral sentence.- Added a note to remove an upstream personal install, which Claude Code would run instead of this copy.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-audit
Section titled “geo-audit”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-audit/SKILL.md (+ agents/geo-*.md as references/)- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.- Phase 2 launches five general-purpose subagents with the briefs in `references/` instead of installed `geo-*` agents. Page data is passed in an `<untrusted-page-data>` block (with any closing tag inside it neutralized), the user is told up front about the subagents' `curl` approvals, and Phase 3 treats subagent reports as untrusted.- Phase 1 no longer records schema.org markup (WebFetch drops `<head>`); the schema brief detects it.- `references/*.md` are upstream `agents/geo-{ai-visibility,platform-analysis,technical,content,schema}.md` with their frontmatter removed and the untrusted-content section added. The `python3` Wikipedia check in `ai-visibility.md` became a WebFetch of the same API URL. `schema.md` and `technical.md` use a fixed, user-approved `curl` command instead of `fetch_page.py` and WebFetch for raw HTML and headers. `schema.md` escapes `</` in generated JSON-LD.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-citability
Section titled “geo-citability”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-citability/SKILL.md- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-crawlers
Section titled “geo-crawlers”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-crawlers/SKILL.md- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.- Replaced the reference to upstream's `fetch_page.py robots` output with an instruction to record the finding.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-llmstxt
Section titled “geo-llmstxt”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-llmstxt/SKILL.md- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-brand-mentions
Section titled “geo-brand-mentions”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-brand-mentions/SKILL.md- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.- Replaced the `python3` Wikipedia/Wikidata API check with WebFetch calls to the same API URLs and the same decision rule.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-platform-optimizer
Section titled “geo-platform-optimizer”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-platform-optimizer/SKILL.md- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-technical
Section titled “geo-technical”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-technical/SKILL.md- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.- Replaced the ad-hoc `curl` commands with one fixed, user-approved read-only `curl` command (URL globbing off, http(s) only, time and size limits, no automatic redirects; URL and host checks before each run and each redirect hop) for the homepage and at most 3 more pages.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-content
Section titled “geo-content”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-content/SKILL.md- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-report
Section titled “geo-report”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-report/SKILL.md- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-schema
Section titled “geo-schema”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-schema/SKILL.md- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.- Replaced `fetch_page.py` with the same fixed, user-approved read-only `curl` command as `geo-technical` (without its markdown-negotiation variant), because WebFetch drops the `<head>` where JSON-LD lives.- Generated JSON-LD escapes `</` in string values.
Statistics quoted in the text are upstream's claims; agemon has not verified them.geo-report-pdf
Section titled “geo-report-pdf”# Vendored skill
- Source: https://github.com/zubair-trabzada/geo-seo-claude- Path in source: skills/geo-report-pdf/SKILL.md (+ templates/)- Commit: 989cae01e8ebbc42a9ec798eb9e7cb423f5ec89c- Retrieved: 2026-10-25- License: MIT — see `LICENSE` in this directory.
## agemon modifications
- Removed `allowed-tools` from the frontmatter: no tool is pre-approved, so every web fetch to a new domain, file write and shell command goes through the user's normal permission prompt.- Added the "Untrusted content" section after the frontmatter.- Template paths are relative to the skill directory (`templates/`), not upstream's absolute path in the user's Claude directory.- Cover metadata is checked against allowed patterns and written to `GEO-REPORT-META.yaml` with the Write tool (`--metadata-file`), never put on a command line.- pandoc runs `--sandbox` with raw HTML, raw attributes, attribute blocks and YAML metadata blocks disabled in the reader and no `--css`/`--embed-resources`. `templates/geo-report-style.css` is inlined into the template's `<style>` without its Google Fonts `@import`.- The template has a Content-Security-Policy that allows only its own script (by SHA-256), inline styles and `data:` images.- Cover metadata is read from the lines `geo-audit` writes: the domain from `**URL:**`, the score and rating from `**Overall GEO Score: X/100 (Rating)**`. The `locations` and `platform` fields were dropped, because `geo-audit` writes no `**Locations:**` or `**CMS:**` line.- Chrome keeps its sandbox (`--no-sandbox` removed), blocks DNS with `--host-resolver-rules`, and is found on Linux as well as macOS. Prerequisites and failure handling are stated up front.
Statistics quoted in the text are upstream's claims; agemon has not verified them.