llms.txt & llms-full.txt

Use llms.txt and llms-full.txt to give AI tools like Cursor, Windsurf, and Grok structured access to X API documentation for accurate code generation.

llms.txt — A structured index of all documentation pages (titles, URLs, short descriptions). Gives AI tools a map of what’s available.

llms-full.txt — The complete documentation in a single Markdown file for maximum context.

The llms.txt standard is like a sitemap for AI. It helps LLMs understand your documentation structure and find relevant content — similar to how sitemap.xml helps search engines.

File	What it contains	Best for
llms.txt	Curated root index + links to section indexes	Quick overview and agent entry point
x-api/llms.txt	370+ X API v2 reference pages	Posts, Users, DMs, Streams, Compliance, etc.
enterprise-api/llms.txt	Enterprise, Account Activity, GNIP, Webhooks	Historical + real-time enterprise data
x-ads-api/llms.txt	Ads API (campaigns, creatives, audiences, analytics)	Advertising and measurement
xdks/python/llms.txt + xdks/typescript/llms.txt	Full SDK client & model references	Python and TypeScript/JavaScript developers
llms-full.txt	Entire docs as one Markdown file	Maximum context for deep reasoning

These files (plus section-specific indexes under /x-api/llms.txt, /enterprise-api/llms.txt, /x-ads-api/llms.txt, and the XDKs) are available at the root and under /.well-known/. Every documentation page supports the .md suffix for clean Markdown (example: https://docs.x.com/x-api/posts/get-post-by-id.md).

---

How to use it

Strongly recommended for agents: Start by reading AGENTS.md (or AGENT.md) for explicit usage guidance.

Most AI tools accept URLs directly. Just provide the URL and the tool will fetch and parse the content:

• Grok: Paste https://docs.x.com/llms-full.txt into the chat and ask questions about the X API
• Cursor / Windsurf: Add https://docs.x.com/llms-full.txt as documentation context for your project
• Custom agents: Fetch the file programmatically and include it in your system prompt or context window

# Fetch the documentation index
curl https://docs.x.com/llms.txt

# Fetch the complete documentation
curl https://docs.x.com/llms-full.txt

# Fetch any individual page as Markdown
curl https://docs.x.com/tools/llms-txt.md