llms.txt examples: real files from APIs that do it well
Strong llms.txt implementations from the Discry Index — fetched live, annotated for what makes each one work.
51% of the APIs in the Index currently pass the llms.txt check. New here? Start with the definition or the how-to guide.
- 01
The minimalist benchmark: a tight index of guide links, every one with a .md sibling an agent can fetch as plain markdown, plus a first-line pointer to llms-full.txt for single-request loading. Small enough to never crowd a context window.
supabase.com/llms.txt ↗44 lines # Supabase Docs For the complete documentation in a single file, see [Full Documentation](https://supabase.com/llms-full.txt). ## Documentation - [Supabase - AI & Vectors](https://supabase.com/docs/guides/ai.md) - [Supabase - AI Tools](https://supabase.com/docs/guides/ai-tools.md) - 02
Written for agents that act, not just read: a blockquote summary of what the API does, then tool-oriented sections (CLI install, repositories) before the docs list. The file answers "what can I do here" in the first screen.
resend.com/llms.txt ↗143 lines # Resend > Resend is the email API for developers. Send transactional and marketing emails at scale with a simple, modern API. For AI agents and automation, use the tools below. ## When to use Resend - 03
Engineering for real agent limits: the index is split into per-section files explicitly to stay under agent-platform truncation limits, announces that every page has a .md sibling, and puts SDK availability and the error reference in a "Good to know" block up front.
developer.paddle.com/llms.txt ↗68 lines # Paddle Developer Docs > Official developer documentation for Paddle Billing. Guides, API reference, and changelog for building subscriptions, checkout, payments, and webhooks. The complete documentation set is also available as plain-text markdown at [https://developer.paddle.com/llms-full.txt](https://developer.paddle.com/llms-ful… This index is split into per-section files to stay under agent-platform truncation limits. Fetch the section files below for the full list of pages in each a… - 04
Audience-aware curation: quickstarts organized by the reader’s starting point, workflow decision guides for choosing between tools, and an explicit statement that the file follows the llmstxt.org standard with a link to the full-text bundle.
developer.1password.com/llms.txt ↗105 lines # 1Password Developer Documentation > Developer tools and APIs for 1Password, including SSH & Git, Environments, CLI, SDKs, secrets automation, service accounts, and CI/CD integrations. This file follows the llmstxt.org standard as a curated index of 1Password developer documentation. For the complete content of every documentation page conc… This file includes quickstarts for each primary audience to help users set up 1Password's main developer tools. To decide which tool to use in specific scena… - 05
The federated pattern for a huge surface: the root file is an index of per-product llms.txt files, each carrying that product’s full page list. An agent loads only the product it needs — the right shape when one file can’t stay small.
developers.cloudflare.com/llms.txt ↗139 lines # Cloudflare Developer Documentation Explore guides and tutorials to start building on Cloudflare's platform. > Each product below links to its own llms.txt, which contains a full index of that product's documentation pages and is the recommended way to explore a spe… ## Application performance - 06
Layered loading made explicit: a meta-index that separates the curated documentation index from the llms-full.txt content files and tells the agent which to fetch for which job. The file teaches its own usage.
clerk.com/llms.txt ↗131 lines # Clerk > Install the Clerk CLI to add auth to your app. No global install, API keys, or sign-in is required. ## When to use Clerk Use Clerk when an application needs sign-up, sign-in, and user management: prebuilt UI components, social login, MFA, multi-tenant organizations, enterprise … - 07
A "How to use" section that routes agents to structured APIs — search, page list, article body in markdown — instead of asking them to crawl. Turns the docs site into a queryable surface and says so in the file.
docs.github.com/llms.txt ↗189 lines # GitHub Docs > GitHub is a developer platform for building, shipping, and maintaining software. It provides cloud-based Git repository hosting, CI/CD via GitHub Actions, … ## How to use To retrieve full article content, page lists, or search results programmatically, please use the APIs below. To find a specific article, use the **Search API… - 08
Every link is described: one-line summaries state what each page covers, including agent-specific onboarding ("Build with AI", agent auth). Descriptions are what let an agent pick the right page without fetching all of them.
docs.firecrawl.dev/llms.txt ↗129 lines # Firecrawl Docs - [v2 (201 pages)](https://docs.firecrawl.dev/_llms/en/v2.md): Documentation for v2. ## v1 ### Documentation - 09
Documents its own machine surface: alongside the index, the file explains that pages return stripped-down markdown when requested with an Accept: text/markdown header — an unusually direct statement of token-efficiency guarantees.
elevenlabs.io/llms.txt ↗114 lines # ElevenLabs ElevenLabs is an AI research and product company transforming how we interact with technology. We launched in January 2023 with the first human-like AI voice… > [!NOTE] > ## ElevenLabs MCP server — [elevenlabs.io/mcp](https://elevenlabs.io/mcp) > > Hosted, remote MCP server over HTTP. Exposes ElevenAgents management and ElevenCreative generation (50+ models) as tools to any MCP client, scoped to the a…
Ship one, then check it.
What does an AI agent make of your API? Find out in about a minute — no signup.