llms-audit
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@llms-auditAudit https://acme.com for AI-answer readiness"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
llms-audit — AI-answer readiness checker
Audit any site for the AI-answer era: llms.txt, schema, sitemap → score + fix list.

Install
npm i -g github:F0Rextasy/llms-auditOffline-friendly: once fetched, all checks run locally against the downloaded HTML/robots/sitemap — no account, no API key, no telemetry.
Related MCP server: maxaeo-ai-visibility-mcp
30-second setup
$ llms-audit https://yoursite.com
score 5/100 — http://127.0.0.1:8790/
check status fix
-------------- ------ ---
llms.txt FAIL Create /llms.txt: first line `# <site name>`, then at least one `## ` section (run with --fix-llms for a template)
llms-full.txt WARN Publish full context at /llms-full.txt (concatenated markdown of key pages)
sitemap.xml FAIL Publish /sitemap.xml with at least one well-formed <url> entry containing <loc>
robots.txt FAIL Allow GPTBot in robots.txt (remove `Disallow: /` for GPTBot)
JSON-LD FAIL Add <script type="application/ld+json"> with @type Organization/WebSite/WebPage/Article
meta tags FAIL Add missing <title>, meta description, canonical to <head>
<h1> PASS -
HTTPS + health FAIL Serve the site over HTTPS and redirect http → httpsFix what it flags, re-run — an AI-ready page (everything present, HTTP local so only HTTPS fails) scores:
$ llms-audit http://127.0.0.1:8791
score 85/100 — http://127.0.0.1:8791/
check status fix
-------------- ------ ---
llms.txt PASS -
llms-full.txt PASS -
sitemap.xml PASS -
robots.txt PASS -
JSON-LD PASS -
meta tags PASS -
<h1> PASS -
HTTPS + health FAIL Serve the site over HTTPS and redirect http → httpsNeed the llms.txt boilerplate? It prints one for your host:
$ llms-audit https://acme.test --fix-llms
# Acme
> One-paragraph summary of Acme for AI answers. Replace this line with what the site is and who it serves.
## Products
- [Example item](https://acme.test/): one-line description of the key offering.Flags: --json (machine report), --out report.html (self-contained dark-mode
report you can attach to a PR), --timeout ms, --fix-llms.
How it works
Eight weighted checks (100 points total), fetched once from the target origin:
check | weight | passes when |
llms.txt | 20 |
|
llms-full.txt | 5 | full-context file exists (advisory) |
sitemap.xml | 15 | parses with ≥1 |
robots.txt | 15 | none of GPTBot, ClaudeBot, Claude-Web, Perplexity-Bot, Google-Extended fully disallowed |
JSON-LD | 15 | parses; |
meta tags | 10 |
|
| 5 | exactly one descriptive heading present |
HTTPS + health | 15 | https final URL, no redirect loop, status < 400 |
Every FAIL/WARN prints the concrete fix, so the report doubles as a worklist.
MCP server
llms-audit --mcp speaks MCP over stdio and exposes audit_url {url} — wire it
into Claude Code, Cursor, or any MCP client:
{"mcpServers": {"llms-audit": {"command": "llms-audit", "args": ["--mcp"]}}}Comparison
llms-audit | open-seo-tool style crawlers | manual checklist | |
Focus | AI-answer readiness (llms.txt, AI bots, JSON-LD) | classic SEO rank factors | you, a notepad |
Output | score + per-check fix, HTML/SARIF-ready JSON, MCP | dashboards, accounts | inconsistent |
Cost | free, offline after fetch | subscription | time |
Regenerate demo
python tools/render_demo.py # spawns fixtures/server.ts (unready + ready), runs the real CLI, rebuilds assets/Fixtures live in fixtures/server.ts; every console block above is a real run.
License
This server cannot be deployed
Maintenance
Related MCP Connectors
Checks llms.txt, AI crawler access in robots.txt, and sitemap - with a 0-100 AI readiness score.
- seegeoOAuthcom.see-geo
Audit any website for AI visibility: graded report, findings with fixes, AI crawler access check.
Generate 18 AI readiness files (llms.txt, ai.txt, RAG indexes, schema) for any website.
Audit any site's AI visibility from your assistant: crawler access, rendering, and schema.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables web audits for performance, security, SEO, and DNS, returning structured improvement prompts for AI assistants.28 npmMIT
- AlicenseAqualityBmaintenanceEnables AI agents to check whether a public website is crawlable, understandable, and ready for AI search workflows through local-only audits of robots.txt, sitemaps, metadata, and llms.txt.350 npm1MIT
- AlicenseAqualityCmaintenanceCheck whether a website is visible to AI search engines (ChatGPT, Perplexity, Claude, Google AI Overviews). Returns a 0-100 readiness score, a grade, and a specific fix for each gap. Dependency-free, no API keys.28 npmMIT
- FlicenseNot gradedqualityDmaintenanceAudits any website for SEO issues, providing scored health checks, schema validation, and performance analysis through AI assistants.-