Item logo image for Agent-Readiness Auditor – AI Agent Website Checker

Agent-Readiness Auditor – AI Agent Website Checker

5.0(

1 rating

)
ExtensionDeveloper Tools7 users
Item media 5 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 1 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 2 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 3 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 4 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 5 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 1 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 1 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 2 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 3 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 4 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker
Item media 5 (screenshot) for Agent-Readiness Auditor – AI Agent Website Checker

Overview

Can AI agents use your website? Score any page for Claude, Gemini & ChatGPT agents, see what they see, and get fixes.

Can an AI agent actually use your website? Agent-Readiness Auditor checks how well agents like Claude, Gemini, ChatGPT agent and Perplexity Comet can read a web page and complete tasks on it, then shows you exactly what to fix, starting with the fixes worth the most points. Open any page and click the icon. The audit starts right away, and in about a second you get a 0–100 score, a breakdown by category, and every problem element with why it matters and how to fix it. The score also appears on the toolbar icon. Most "agent-ready" checkers only look for files like robots.txt and llms.txt. This one checks whether an agent can find your controls by name, fill in your forms, get past overlays, buy from your product pages and read your content without being misled. NEW IN 1.6 ★ Accessibility overlap: checks that are also WCAG 2.2 failures say so, such as unnamed buttons (4.1.2), missing alt text (1.1.1) and keyboard access (2.1.1), with a WCAG filter and links to W3C's guidance. Fix once, and it helps people using screen readers and keyboards as well as AI agents. ★ Run the steps: after an Agent test, watch the plan carried out on the live page in a background tab, step by step, with what happened at each one. It never pays, places orders, deletes anything or types personal details. NEW IN 1.5 ★ Weekly re-scan with alerts: after a site scan, turn on "Re-scan every week". The same pages are audited again each week, and you get an alert if the average score drops or a new problem spreads across pages. ★ Keyboard and focus checks: controls you can't reach with the Tab key, and controls with no visible focus indicator. ★ Shadow DOM and frames: buttons and fields inside web components (open shadow roots) and same-site iframes are now audited too. ★ Name suggestions from on-device AI: for an unnamed button or image, Chrome's built-in AI (Gemini Nano) suggests a name from the text around it. Nothing leaves your computer. ★ Command line and MCP server: run the same audit in CI with "npx agent-readiness-audit <url> --min-score 80", or let Claude Code and Cursor audit pages themselves through MCP. ★ In 6 languages: English, Español, Deutsch, 日本語, Português (Brasil) and বাংলা. NEW IN 1.4 ★ Site scan, free: audit up to 25 pages from your sitemap or navigation and see which problems repeat across pages. A problem on most pages comes from one shared template, so one fix covers them all. Copy one fix prompt for the shared problems, open a site report, or download a CSV. ★ Suggested code for every problem element: a real button instead of a clickable div, a label for a field, a name for an icon button guessed from the page. Native HTML first, ready to copy. ★ Compare with a competitor: put your page next to any page you've audited, with scores by category and the checks where each one does better. ★ Re-run on reload: while you edit your site, the audit runs again every time the page reloads. ALSO NEW (1.3) ★ Agent test: type a task like "Add this product to the cart". Chrome's built-in AI (Gemini Nano) reads your page the way agents do and plans the clicks, then the auditor shows each step on the page and where an agent would get stuck, such as an "Add to cart" that looks like a button but isn't one. It runs on your computer: free, private, no API key. On computers without Chrome's built-in AI, the test matches the task to control names. ★ Fix these first: the three fixes worth the most points, with the score you'd reach, and one prompt that covers them. ★ Mark issues on page: numbered markers on every problem element at once. Hover a number for details. ★ Shareable report: a clean report to send to a client or teammate, or save as PDF. ★ Shopping agents: on product pages, checks Product structured data, price and stock, shipping, returns and ratings, and a clear "Add to cart" button. ★ Try it in seconds: a demo store with problems planted on purpose, and your recent audits on the start screen. ALSO INCLUDED ★ Agent view: your page as an agent reads it, as its accessibility tree in text, with a token estimate. Unnamed buttons and links are flagged in red. ★ Prompt-injection check: finds hidden text aimed at AI agents, such as "ignore previous instructions" in invisible or off-screen text, aria-labels or HTML comments. ★ Fix prompts for coding agents: copy the report, the top fixes or one issue as a ready-made task for Claude Code, Cursor or Copilot, with a fix, test and verify loop. ★ Changes since your last audit: run it again after a fix to see your score change and which issues are fixed, new or reduced. WHAT IT CHECKS (42 checks) • Real controls: clickable divs and spans that aren't real buttons or links, links that go nowhere, controls the Tab key skips, no visible focus, positive tabindex • Labels and names: icon-only buttons with no label, vague names like "Click here", images without alt text, misleading ARIA • Forms: unlabeled or placeholder-only fields, missing input types and autocomplete hints, required fields not marked as required • Interaction traps: cookie banners and overlays, unnamed dialogs, CAPTCHAs, cross-origin frames, shadow DOM and same-site frames, tiny click targets, layout shifts • Stable targeting: controls that can only be found by generated class names or auto-generated IDs • Page structure: title, language, landmarks, headings, page size, content that needs JavaScript, where the main content starts • Shopping agents (product pages): Product data, price and availability, shipping, returns, ratings, the cart button • Agent safety: hidden instructions aimed at AI agents • Machine-readable signals: meta description, Open Graph, canonical link, structured data, llms.txt, robots.txt rules for AI • Agent protocols (for information, not scored): Markdown for agents, sitemap, MCP server card, API and OAuth discovery, Content Signals, WebMCP WHO IT'S FOR Front-end developers, QA and accessibility teams, SEO and GEO specialists, e-commerce teams, and agencies preparing sites for AI agents and agentic browsers. Agencies: scan a client's site, compare it with competitors, and send the report. PRIVATE BY DESIGN Audits run entirely in your browser. No account, no servers, no analytics, no tracking. Nothing about the page is sent anywhere, and the Agent test's AI runs on your own computer. The extension only reads a page when you click its icon, and asks for access to a site only when you start a site scan or turn on re-run on reload. Weekly re-scans run in the background only for the sites you turn them on for, and alerts only appear if you allow notifications.

Details

  • Version
    1.6.0
  • Updated
    October 30, 2026
  • Offered by
    sabyasachi.mondal1@gmail.com
  • Size
    316KiB
  • Languages
    6 languages
  • Developer
    Email
    sabyasachi.mondal1@gmail.com
  • Non-trader
    This developer has not identified itself as a trader. For consumers in the European Union, please note that consumer rights do not apply to contracts between you and this developer.

Privacy

Manage extensions and learn how they're being used in your organization
The developer has disclosed that it will not collect or use your data. To learn more, see the developer’s privacy policy.

This developer declares that your data is

  • Not being sold to third parties, outside of the approved use cases
  • Not being used or transferred for purposes that are unrelated to the item's core functionality
  • Not being used or transferred to determine creditworthiness or for lending purposes

Support

For help with questions, suggestions, or problems, please open this page on your desktop browser

Search
Clear search
Close search
Google apps
Main menu