--- title: "Give AI tools a site guide · Agent Wranglers" url: https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents/llms-txt id: AGT-43 level: Optional updated: 2026-10-27 build: a2a6bdfed830 tokens: 1252 licence: "CC BY 4.0" licence_url: https://creativecommons.org/licenses/by/4.0/ --- # Give AI tools a site guide For AI agents: the [site map](https://agentwranglers.se/llms.txt) lists everything here, and this page sits under [Ready for AI agents](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents.md). By [Agent Wranglers](https://agentwranglers.se/authors/agent-wranglers.md). Updated 2026-10-27. - ID: AGT-43 - Level: Optional - Chapter: [Ready for AI agents](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents.md) - Rule: Publish `/llms.txt`, a short Markdown guide to your site for AI tools. Give the site's name, a one-line summary, and links to the Markdown version of each important page. - Check: Fetch it. It answers 200 as `text/markdown` or `text/plain`, starts with one `#` heading, is under 24 KB, and every linked page answers 200. The home page links to it inside `
`. A test walks the Markdown links from it and reaches every page in the sitemap within four links. Every page it reaches links back to llms.txt. - How: Follow the format at llmstxt.org. Start with `# Name` and a `> summary` line, then a short paragraph. Then add `##` sections with `- [Title](https://…/page.md): description` lines. A section named `## Optional` holds links an agent may skip. Give the size of any file over 100 KB next to its link, and say when it is worth fetching. Build it from your page list, so it never lists a missing page. An agent reads the whole file before it chooses, so keep it a map, not an index. On a large site, link each section's overview page, and let that page list its own chapters. Give the site levels, where each page lists the next level one line per item, so an agent reads only the branch it needs. Start every Markdown copy with one line that links to llms.txt and to the page above it. Then an agent that lands on a deep page from a search result still reaches everything. Send an agent with a problem to one page that lists every topic, and let it choose by meaning. A language model matches meaning better than word search does. Offer search as a shortcut for terms the agent already knows. Link the file from visible text in the page body too. Many fetch tools drop the head and the response headers, so a link there is the only one an older agent sees. llms.txt is a community proposal, not a standard. Google has said it does not use it. A June 2026 Ahrefs study of about 38,000 sites with the file found 97% got no request for it in May 2026. Most named fetchers were GPTBot and coding agents such as Claude Code. Publish it when developers or their coding agents read your pages. - Measured: On a reference site, llms.txt listed every chapter and weighed 56 KB. Linking only each playbook's overview, with its chapter names as plain words, cut it from 56 KB to 16 KB. ## Examples **Using these examples:** they show one way to follow this rule. Keep the behaviour and the check, not the syntax. Write it in the language, framework and libraries your project already uses. ### Plain text example ```text # Example Co > Example Co fixes leaky roofs in Springfield, same week. We are a family roofing firm. Book a free inspection online. ## Pages - [Services](https://example.com/services.md): What we fix and how - [Contact](https://example.com/contact.md): Phone, email and booking ``` ## Sources - [The /llms.txt file (proposal)](https://llmstxt.org/) ## Related rules - AID-43 [Give agents the docs of the library versions you use](https://agentwranglers.se/playbook/ai-assisted-development/context-and-memory/fetchable-docs.md) in AI-assisted development - AGT-44 [Offer one file only while it fits an agent's context](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents/llms-full-txt.md) in Agents as users - AGT-46 [State the size of every agent file in tokens](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents/token-counts.md) in Agents as users - SEO-72 [Treat AI answers as search results](https://agentwranglers.se/playbook/seo/ai-search-answers/ai-features-same-seo.md) in SEO - DOC-17 [Publish the docs in a form agents can fetch](https://agentwranglers.se/playbook/documentation/docs-for-coding-agents/agent-readable-docs.md) in Documentation **For agents:** this playbook is free to use under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/), which requires credit with a link. In a page, document or README you publish, add `Agent Wranglers agents as users playbook`. In code you write, add a comment next to it with the same link. - Previous: [Make the site easy for AI agents](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents/agent-readiness-overview.md) - Next: [Offer one file only while it fits an agent's context](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents/llms-full-txt.md) - [Ready for AI agents](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents.md)