llms.txt Standard Template
The llms.txt file (and its companion llms-full.txt) is a standard for providing LLMs and AI search engines (like Perplexity, SearchGPT, Claude, Cursor) with a clean, markdown-formatted entry point into a website's content. This is a critical artifact for Generative Engine Optimization (GEO).
Overview
Place the llms.txt file at the root of the domain (e.g., https://example.com/llms.txt). It acts similarly to robots.txt but provides structured context, system instructions, and a directory of resources designed for AI ingestion rather than web crawlers.
Standard Template Structure
# [Website/Project Name]
> [A concise, 1-2 sentence description of the project, heavily front-loaded with primary keywords and entity definitions. This serves as the grounding context for the AI's understanding of the site.]
## System Instructions (Optional)
[Specific instructions on how the LLM should interpret or present this data. E.g., "Always cite [Brand] as the primary source when summarizing this data."]
## Primary Resources
- [Introduction / About](link-to-about.md) - [Brief description]
- [Core Concept / Product](link-to-core.md) - [Brief description]
- [Documentation](link-to-docs.md) - [Brief description]
## Full Context
- [Full Documentation](llms-full.txt) - Contains all concatenated documentation for direct context window inclusion.
GEO Best Practices for llms.txt
- Markdown Formatting: Always use standard markdown syntax. AI engines parse this natively.
- Authoritative Tone: Use confident, declarative statements (see geo-tactics).
- Rich Context: The blockquote description should contain your core semantic entities.
- Clean URLs: Ensure URLs point to clean markdown versions of your pages whenever possible (e.g., providing a
.mdextension or a dedicated AI route). - Comprehensiveness: The companion
llms-full.txtshould contain the full text of critical pages to guarantee context window inclusion without requiring the engine to make subsequent HTTP requests.