llms-txtlisted
Install: claude install-skill visibian/seo-geo-skills
# llms.txt
Write or check an llms.txt file, and tell the user plainly what it does and does not do.
## What to tell the user first
- llms.txt is a proposal (llmstxt.org, Jeremy Howard, first published 3 September 2024, now in its second version). It is a short markdown file that guides agents to a site's key pages.
- It is used most by coding agents reading software documentation. The AI labs publish llms.txt files for their own developer docs, and Chrome's Lighthouse checks for one in its agentic browsing audit (llmstxt.org, v2).
- Google Search does not use it. Google's guide to generative AI features (updated 10 July 2026) lists llms.txt among the things to ignore for Google Search, and says it "will neither harm nor help" a site there.
- No major AI assistant has documented that it reads llms.txt to answer questions about a brand. Say so, and present the file as a low-cost aid for agents and documentation, not a visibility lever.
Sources and dates: `references/sources.md`.
## Steps
1. Check for an existing file. From this skill's folder: `python3 scripts/check_llms_txt.py example.com` (or `--file path/to/llms.txt` in a codebase). Add `--links` to find links that no longer open.
2. Gather the key pages. Read the sitemap, the main navigation and the footer, or the codebase's routes. Pick the pages an agent would need to answer questions about the business: what it is, products and prices, docs, policies, contact. Keep the list short.
3. Write the file in the order the