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 agentsAI Search Auditor: Get Found and Cited by AI Answers
Checks whether ChatGPT, Claude, Perplexity and Google AI can fetch, read and quote your pages, shows the evidence, and writes the fixes with your own facts. Includes a Python checker.
$14
AI Search Auditor: Get Found and Cited by AI Answers
Example session with this skill installed
Run an AI search audit on https://example.com/ and tell me what to fix first.
- Read your context and instructions
- Compiled the ai search auditor
Summary
example.com scores 49/100. Every AI crawler may fetch it and no firewall gets in the way, but the page gives an answer engine almost nothing to quote: 27 words, no H1, no structured data, no sitemap.
Checker output (geo_check.py)
AI search readiness: 49/100 (https://example.com/)
Crawler access 30.0 / 30
Indexability 6.0 / 15
Rendering 12.0 / 25
Structured data 0.0 / 15
Trust 0.0 / 10
AI-native files 0.0 / 3
Fix first (points lost):
- 7.0 [WARN] Main text is in the HTML: 27 words in the raw HTML, 1 script tag
- 5.0 [WARN] Sitemap: https://example.com/sitemap.xml answered 404
- 5.0 [FAIL] JSON-LD present and parses: 0 blocks
- 5.0 [WARN] A page type (Article, Product, FAQPage, Service...): none
- 5.0 [WARN] Who is behind the site (Organization or Person, with sameAs): none
- 4.0 [WARN] Canonical URL: canonical=none
- 4.0 [WARN] One H1 and a heading outline: H1 x0, H2 x0
- 4.0 [WARN] A visible or structured date: none found
Top fixes, in order
| # | Fix | Evidence | Effort |
|---|---|---|---|
| 1 | Add an H1 and a 40 to 80 word opening that says what the domain is for | 27 words, no H1 | S |
| 2 | Add Organization JSON-LD with the owner's name, url and sameAs profiles | 0 JSON-LD blocks | S |
| 3 | Add a self-canonical tag and a sitemap named in robots.txt | no canonical, sitemap 404 | S |
| 4 | Show a "last reviewed" date and mirror it in dateModified | no date on the page | S |
Blockers: none. Crawler access is already 30/30, so the work is all content and markup.
Next steps
- Re-run
python3 scripts/geo_check.py https://example.com/after the changes and compare the score. - Hand the target questions to ai-visibility-tracker to set a baseline before anything ships.
Connects securely to your tools. The creator never sees your data.
About this skill
Most AI SEO advice is a checklist you could have guessed. This runs a real check first, shows you the evidence for every finding, and writes the fixes with the facts already on your site.
What it finds
- Which of 18 AI and search crawlers your robots.txt lets in, read with Google's matching rules (most specific group, longest rule, wildcards). Answer crawlers and training crawlers are kept apart: blocking GPTBot does not take you out of ChatGPT search, blocking OAI-SearchBot does.
- A firewall that answers AI crawler user agents with a block page while robots.txt says yes.
- Pages whose main text only appears after JavaScript runs, which most AI crawlers never see.
- noindex, canonical, sitemap, headings, the opening paragraph, JSON-LD, dates, authors, outbound sources and llms.txt.
What you get back
- A score out of 100 with the costliest problems first, and every check saved as JSON.
- A fix pack: robots.txt lines, an llms.txt built from your real pages, JSON-LD filled only with facts from the page, and answer blocks rewritten so an engine can quote them.
- A monthly tracker that scores neutral prompts on their own. Prompts that name your brand almost always mention your brand, so counting them makes visibility look better than it is.
What is inside
- 3 skills: ai-search-audit, ai-visibility-tracker and citable-rewrite
- geo_check.py: one Python file, no dependencies, no API keys
- 2 reference files (AI crawlers and what blocking each one does; fix snippets) and 2 templates (a 20-prompt set and a CSV log)
Use cases
- Find out why assistants recommend competitors and never you.
- Check a site before and after a redesign or a move to a JavaScript framework.
- Decide what to do about AI crawlers in robots.txt without losing search visibility.
- Report AI visibility to a client or a boss with numbers that hold up.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Marketing & SEO.
One URL per run. No rank tracking or backlink data. Visibility runs use the assistants you already have. The firewall test sends four requests with crawler user agents, so use it only on sites you own or may test. No tool can promise citations: this removes what blocks the engines and measures the change.
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
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 5 days ago
- Software
Python 3.8+ (for the checker script)
Permissions
File Scopes
Terminal runs scripts/geo_check.py. Network access fetches the page you audit plus its robots.txt, sitemap, llms.txt and two /.well-known files (about 12 requests per URL). Write access only for the optional JSON report (--out).