More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Make Your Site Readable to AI Search: 40-Check Audit

    1

    A single line in robots.txt can block GPTBot, ClaudeBot or PerplexityBot; a leftover noindex hides the page; no structured data means no clear facts to cite.

    $9.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Make Your Site Readable to AI Search: 40-Check Audit

    Make Your Site Readable to AI Search: 40-Check Audit

    Example session with this skill installed

    Audit our bakery page sample/bad.html for AI search visibility. Use sample/bad-robots.txt as its robots file. Give me the score, every PASS/WARN/FAIL with the exact fix, then compare with the fixed version.

    • Read your context and instructions
    • Compiled the make your site
    • Generated the UI component

    AI search visibility audit: sample/bad.html (10 Nov 2026)

    Verdict: BLOCKED Score: 28.6/100

    Checks

    3 PASS | 14 WARN | 18 FAIL | 5 N/A. Score working: (3 + 0.5 x 14) / 35 x 100 = 28.6. N/A checks are excluded.

    Findings

    IDStatusElement or fileMeasured valueExact fix
    CA1FAILrobots.txtblocked; Disallow: /Add a GPTBot group with Allow: / or remove the block.
    CA2WARNrobots.txtOAI-SearchBot not mentionedAdd an explicit OAI-SearchBot policy.
    CA3WARNrobots.txtChatGPT-User not mentionedAdd an explicit ChatGPT-User policy.
    CA4FAILrobots.txtblocked; Disallow: /Add a PerplexityBot group with Allow: / or remove the block.
    CA5FAILrobots.txtblocked; Disallow: /Add a ClaudeBot group with Allow: / or remove the block.
    CA6WARNrobots.txtClaude-SearchBot not mentionedAdd an explicit Claude-SearchBot policy.
    CA7FAILrobots.txtblocked; Disallow: /Add a Google-Extended group with Allow: / or remove the block.
    CA8WARNrobots.txtBingbot not mentionedAdd an explicit Bingbot policy.
    CA9WARNrobots.txtApplebot-Extended not mentionedAdd an explicit Applebot-Extended policy.
    CA10PASSrobots.txtfile read successfullyNo change needed.
    CA11FAILmeta[name=robots]noindex, nosnippetRemove noindex.
    CA12FAILmeta[name=robots]noindex, nosnippetRemove nosnippet or max-snippet:0.
    CA13N/AHTTP headerslocal file modeRe-run against the deployed URL.
    PB1WARNtitle16 charactersWrite a descriptive 30-60 character title.
    PB2FAILmeta description0 charactersAdd a specific 70-160 character summary.
    PB3WARNcanonical0 tagsAdd exactly one absolute canonical URL.
    PB4FAILH12 foundUse exactly one visible H1.
    PB5PASSheading sequenceno skipped levelsNo change needed.
    PB6FAILhtml[lang]missingAdd the page language.
    PB7FAILOpen Graph tags0/3 presentAdd og:title, og:description and absolute og:image.
    PB8FAILbody text40 wordsAdd useful page-specific text to reach at least 300 words.
    PB9WARNraw HTML body40 visible wordsServer-render the main answer text.
    PB10FAILimage alt1/1 missingAdd concise alt text.
    PB11FAILinternal links0 foundAdd at least three relevant internal links.
    SD1FAILJSON-LD block0 blocksAdd JSON-LD for the primary entity.
    SD2N/AJSON-LD parserno blocksAdd JSON-LD, then validate its syntax.
    SD3FAILJSON-LD @typemissingAdd a recognized schema.org @type that accurately matches the page entity.

    bad.html

    TSX · React component

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    About this skill

    AI search tools can only quote what they can reach and read. A single line in robots.txt can block GPTBot, ClaudeBot or PerplexityBot; a leftover noindex hides the page; no structured data means no clear facts to cite. This skill turns your agent into an auditor that measures these things instead of guessing.

    It runs 40 checks in 6 groups: crawler access, page basics, structured data, answer-ready content, trust and freshness, and site files (llms.txt, sitemap).

    The proof

    A real run on a sample local-bakery page (the files ship inside the skill, so you can check them):

    • Before: 28.6/100, 18 FAIL and 14 WARN: four AI crawlers blocked, noindex and nosnippet set, no JSON-LD, no meta description, two H1s.
    • After the fixes: 100/100, 0 FAIL, 0 WARN across 38 checks, re-checked by the same script.

    What you get

    • SKILL.md for Claude Code and OpenAI Codex
    • CHECKS.md: 40 checks, each with a threshold, the reason and a public source
    • geo_audit.py: standard-library Python, nothing to install, audits a live URL or a local file
    • A report format: PASS / WARN / FAIL per check, the element, the measured value and the exact fix
    • Sample pages: a faulty page, its fixed version, robots files, an llms.txt and the real outputs
    • 21 unit tests, testing notes and a changelog

    How to use

    1. Ask your agent: "Audit https://yoursite.com/page for AI search visibility" (or point it at a local HTML file).
    2. Read the scored report and let your agent apply the fixes.
    3. Ask it to re-check until the page is clean.

    Known limitations

    • It cannot see whether an AI engine actually cites your page; it checks everything those engines can read.
    • Not a ranking guarantee.
    • One page at a time (plus that site's robots.txt, llms.txt and sitemap).
    • Needs Python 3.9+.

    FAQ

    Does it change my site? No. It only reads the page and the site files. Your agent applies fixes only if you ask.

    Do I need API keys or accounts? No. It sends nothing anywhere except the requests to the page you audit.

    Does it work on any site builder? Yes. It reads the public HTML, robots.txt, llms.txt and sitemap, whatever made them.

    Mac or Linux? Yes. Tested on Windows 11; it is standard-library Python.

    Requires Python 3.9+. Does not render JavaScript or bypass authentication. Reports on visibility factors, not guaranteed rankings or citation counts.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    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 7 days ago

    Requires

    Python

    Permissions

    Terminal / Shell
    Network Access
    Read Files
    Write Files

    Allowed Hosts

    example.com
    Listed7 days ago

    What's inside

    Frequently Asked Questions

    A classic SEO audit looks at keywords and backlinks. This audit checks what AI search tools need to reach and read a page: crawler access in robots.txt, noindex and snippet rules, structured data (JSON-LD), clear headings and direct answers, author and dates, and site files such as llms.txt and the sitemap.

    It checks the access rules for the crawlers these tools publish, including GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-SearchBot, PerplexityBot, Google-Extended, Bingbot and Applebot-Extended, plus the page signals they can read. It does not query the engines themselves.

    No complex setup is required beyond a Python 3.9+ environment; the skill works instantly with any public URL or local HTML file for immediate analysis.

    You receive the full Python-based auditing tool, the library of 40 specific technical checks, and the reporting engine that generates the 0-100 score and prioritized fix lists.

    This skill focuses on raw HTML and metadata; it does not render client-side JavaScript or bypass login screens, so it is best used on server-side rendered pages or static exports.

    It can remove the things that stop AI search reading or quoting a page, such as blocked crawlers, noindex or nosnippet, missing structured data, and answers buried in long text. It cannot see or promise whether an engine will cite you; no tool can guarantee that.

    New in Marketing & SEO