airready

Is your website ready for AI agents and LLM crawlers? airready grades any site on the signals that decide whether ChatGPT, Claude, Perplexity & co. can find, read, and cite it — and gives you a score, a grade, and the fixes.

pipx install mujin-airready
airready scan https://example.com

AI-readiness: https://example.com
  [----] +25 llms.txt: no /llms.txt
         → Add an /llms.txt summarizing your site for LLMs — see llmstxt.org.
  [----] +15 structured data (JSON-LD): no JSON-LD
         → Add schema.org JSON-LD so machines understand your content.
  [PASS] +8  page title: title present
  ...
  Score: 48/100  (grade D)

Gate it in CI

airready scan exits non-zero when the site falls below a bar, so a nightly job or a pre-deploy step can guard AI-readiness like any other check:

airready scan https://yoursite.com --min-grade B   # or --min-score 75

Wrapped as a GitHub Action: airready-action.

What it checks

Web tool & badge (self-hosted on Vercel)

This repo also ships the AI-Readiness Grader web tool: a URL box that grades any site and gives you an embeddable "AI-Ready" badge — paste it on your site and it links back to your full report. (/api/scan, /api/badge, index.html; deploys on Vercel zero-config; the functions are self-contained and stdlib-only, so the deploy has no dependencies.)

<a href="https://airready.mujinlabs.com/?url=YOURSITE">
  <img src="https://airready.mujinlabs.com/api/badge?grade=A&score=92" alt="AI-Ready: A">
</a>

Built by Mujin Labs — tooling for the autonomous-agent era. MIT.