Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
DeepSeek
Muse+22 moreClaude CodeCursorCodex CLIDeepSeekMuse
Dots
Manus
Grok
Grok Bot+18 agentsGEO-SEO Suite — AI Search Optimization
Full suite of 15+ GEO+SEO audit tools that scores your visibility across ChatGPT, Perplexity and Google AI Overviews 0-100 and ships the client-ready report.
$9.99
GEO-SEO Suite — AI Search Optimization
Example session with this skill installed
/geo quick https://electron-srl.com
- Read your context and instructions
- Compiled the geo-seo suite
Summary
Live GEO audit of electron-srl.com (B2B manufacturer of educational equipment, founded 1991,
exports to 70-80 countries). GEO Score: 28/100 — CRITICAL. The company is real, established
and has a solid product, but it is effectively invisible to AI search engines. Three blocking
issues account for almost all of the loss.
GEO Score: 28/100 — CRITICAL
| Category | Score | Weight | Weighted | Status |
|---|---|---|---|---|
| AI citability & visibility | 18/100 | 25% | 4.5 | CRITICAL |
| Brand authority signals | 15/100 | 20% | 3.0 | CRITICAL |
| Content quality & E-E-A-T | 35/100 | 20% | 7.0 | POOR |
| Technical foundations | 38/100 | 15% | 5.7 | POOR |
| Structured data | 8/100 | 10% | 0.8 | CRITICAL |
| Platform optimization | 35/100 | 10% | 3.5 | POOR |
| TOTAL | 24.5 → 28/100 | CRITICAL |
AI platform readiness
| Platform | Score | Main gap |
|---|---|---|
| Google AI Overviews | 25/100 | No structured data, no Q&A content, catalog-only pages |
| ChatGPT Web Search | 18/100 | No Wikipedia/Wikidata, no schema, 403 to non-browser agents |
| Perplexity AI | 20/100 | No Reddit presence, no original research, no citations |
| Google Gemini | 12/100 | No YouTube channel, no Knowledge Panel, no sameAs links |
| Bing Copilot | 28/100 | No LinkedIn page, no IndexNow, no Bing Webmaster Tools |
Critical finding 1 — Site returns 403 to non-browser user agents
Evidence: every fetch attempt (Python requests, WebFetch) returned HTTP 403 Forbidden. Only
browser-header requests and Google's crawler get through.
Impact: GPTBot, ClaudeBot, PerplexityBot and CCBot all receive 403. The site's content
cannot enter any AI-generated answer, no matter how good it is. This alone explains the 18/100
citability score.
Fix (priority 1, ~30 minutes): add WAF/CDN exceptions for legitimate AI crawler user agents,
verified against their published IP ranges.
Critical finding 2 — Zero structured data
Evidence: 0 JSON-LD blocks, 0 microdata, 0 RDFa across the homepage, /about/ and
/contact-us/.
Impact: no AI system can resolve "Electron Srl" as an entity — not its headquarters, not its
founding year, not its catalog, not its social profiles. That's why Gemini scores 12/100.
Fix (priority 1, ~2 hours): deploy the JSON-LD generated below.
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "Electron Srl",
"foundingDate": "1991",
"url": "https://electron-srl.com",
"address": {
"@type": "PostalAddress",
"streetAddress": "Via Massimo D'Antona 6T",
"addressLocality": "Chiaravalle",
"addressRegion": "AN",
"postalCode": "60033",
"addressCountry": "IT"
},
"sameAs": [
"https://www.linkedin.com/company/electron-srl",
"https://it.wikipedia.org/wiki/Electron_Srl"
]
}
Prioritized action plan
| # | Action | Impact | Effort | Est. points |
|---|---|---|---|---|
| 1 | Unblock AI crawlers at the WAF | Critical | 30 min | +14 |
| 2 | Deploy Organization + Product schema | Critical | 2 h | +9 |
| 3 | Publish llms.txt at the root | High | 1 h | +4 |
| 4 | Rewrite 10 key pages with 134-167 word answer blocks | High | 8 h | +11 |
| 5 | Create a company LinkedIn page + 3 technical posts | Medium | 3 h | +6 |
| 6 | Add author credentials and bylines to the technical section | Medium | 4 h | +5 |
Projection: 28/100 → 77/100 in 90 days by executing all six.
Next steps
- Run
/geo audit https://electron-srl.comfor the full 50-page analysis - Run
/geo report-pdffor the branded client-ready PDF - Run
/geo proposal electron-srl.comto generate the priced commercial proposal - In 30 days,
/geo compare electron-srl.comfor the progress report
Connects securely to your tools. The creator never sees your data.
About this skill
Your site is built for Google. AI search has different rules.
Only 11% of domains get cited by both ChatGPT and Google AI Overviews for the same query. Every AI search engine uses its own index, ranking logic and source preferences. A page ranking #1 on Google can be invisible to Perplexity. Meanwhile, traffic from AI answers already converts 4.4x better than traditional organic.
GEO-SEO Suite audits that gap and hands you the plan to close it.
How the audit works
Give it a URL and your agent runs a parallel audit across six dimensions. You get a composite GEO Score from 0 to 100, the per-category breakdown, and a prioritized action plan with quick wins.
- AI visibility and citability (25%): extractable passages, answer blocks, crawler access
- Brand authority signals (20%): mentions on Reddit, YouTube, Wikipedia, LinkedIn
- Content quality and E-E-A-T (20%): experience signals, original data, author credentials
- Technical foundations (15%): SSR, Core Web Vitals, crawlability, security
- Structured data (10%): JSON-LD schema, validation, rich result eligibility
- Platform optimization (10%): Google AI Overviews, ChatGPT, Perplexity, Gemini, Copilot
15 playbooks, one package
/geo audit: full audit with 5 subagents running in parallel/geo quick: 60-second visibility snapshot/geo citability: scores passages against the Princeton and Georgia Tech research on what LLMs cite (134 to 167 words, self-contained, data rich)/geo crawlers: checks whether your own robots.txt blocks AI crawlers. Over 35% of the top 1,000 sites accidentally block at least one major AI crawler/geo llmstxt: analyzes or generates your llms.txt file/geo brands: finds unlinked brand mentions, which correlate 3x more strongly with AI citation than backlinks (Ahrefs, 75,000 brands analyzed)/geo platforms: per-platform checklist and scoring/geo schema: detects, validates and generates the JSON-LD you are missing/geo technical: technical SEO focused on SSR, because AI crawlers do not execute JavaScript/geo content: E-E-A-T evaluation against the Quality Rater Guidelines/geo reportand/geo report-pdf: client deliverable in Markdown and PDF, with cover page, color-coded score tables and a 90-day roadmap
Built for selling GEO as a service
/geo prospect: lightweight CRM from lead to won or lost, with audit history, notes and deal values/geo proposal: generates the service proposal straight from audit data. Translates technical gaps into business pain points, with three pricing tiers and an ROI projection/geo compare: monthly progress report. The strongest retention tool in the package: it shows a client exactly how many points they gained
White-labeling included. Put your logo, colors and footer on every deliverable by editing one JSON file. The PDF templates are plain HTML and CSS.
Who it is for
- SEO agencies and freelancers selling GEO as a service, delivering in hours instead of days
- In-house marketing teams that need to know whether their brand shows up in AI answers
- Consultants who want a professional deliverable from the first meeting
How it is built
A single root SKILL.md acts as a router and loads only the playbook it needs. Progressive disclosure keeps your context window small. Ships with 5 subagent definitions for parallel analysis, 5 Python utilities, 6 paste-ready JSON-LD templates and full reference documentation.
Works with zero installs. The core audits run on WebFetch and curl. Python, pandoc and Chrome are optional; they add speed or enable the PDF.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills to Reverse Engineer and Scrape Websites.
Crawl limit is 50 pages per audit. PDF generation requires local installation of Pandoc and Chrome. Subagent execution depends on the user environment's support for parallel processing.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
This skill has been used 125 times.
Trust & safety
Security scan
Security scan passed 1 month ago
- Accounts
Google Account
LinkedIn
GitHub
OpenAI
Anthropic
Youtube
Reddit
Crunchbase
Wikidata
G2
Producthunt
Quora
Trustpilot
Perplexity
Bing
- Software
Python
Permissions
Allowed Hosts