llms.txt examples

How real projects summarise themselves for language models.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Worked for most · soon

Vercel Hiding risky files

Vercelllms.txt

dev3000 / www

vercel-labs/dev3000/www/llms.txt

A managed browser and a bounded server/browser event timeline for coding agents. d3k is written in TypeScript and runs under Node.js 24+. Install its skill with pnpm dlx skills add vercel-labs/dev3000 --skill d3k, then ask your agent to run and debug your project. No global CLI installation is required for skill use. The same runtime backs d3k, dev3000, and d3k-runtime. Commands: - start: own the dev server and project browser in a retained process - status: report readiness, session ID,…

1.6k7mo agoReviewMount

Vercelllms.txt

arcade / site

vercel-labs/arcade/apps/site/public/llms.txt

Arcade is an open-source TypeScript CPU 3D renderer, retained-mode terminal UI toolkit, and agent-playable game harness. Canonical site: https://ascii-arcade.dev Repository: https://github.com/vercel-labs/arcade Docs: https://ascii-arcade.dev/docs Using Arcade: https://ascii-arcade.dev/docs/app Controls and tutorial: https://ascii-arcade.dev/docs/app/controls Models, teams, and billing: https://ascii-arcade.dev/docs/app/models Browser integration: https://ascii-arcade.dev/docs/web Hosted CLI: https://ascii-arcade.dev/docs/browser-host HTTP interface: https://ascii-arcade.dev/openapi.json Capability status: https://ascii-arcade.dev/api/v1/status Use Arcade when you need CPU-rendered 3D graphics in a terminal or browser, retained terminal UI components, presentation-independent board-game rules, deterministic visual snapshots, or a persistent self-play harness for agent-driven games. Do not…

212mo agoReviewMount

Vercelllms.txt

tw-blocks / apps

vercel-labs/tw-blocks/apps/docs/public/llms.txt

No summary in the file. Open it to read it.

18mo agoReviewMount

CLAUDE.md vs AGENTS.md

About llms.txt

What is llms.txt?

A markdown file at a website's root that gives language models a short guide to the site and links to the pages worth reading.

What's llms-full.txt?

The same idea with the content included, so an agent can read the documentation in one request.

Who reads it?

Agents and tools that fetch documentation on someone's behalf. It's a proposal, not a standard, and support varies.

How do I write one?

Start with a one-line summary, then sections of links with a sentence each. The examples here show what real projects do.