← Docs

llms.txt Protocol

Site navigation for AI agents.

What is llms.txt?

llms.txt is a convention for helping AI agents navigate websites. Like robots.txt tells crawlers what to index, llms.txt tells agents what content exists and how to find it.

Learn more at llmstxt.org →

Our Implementation

Complyant provides llms.txt at two levels:

Global Index

GET https://complyant.ai/llms.txt

Returns a complete site overview: what regulations exist, how to access different content types, topic routing guidance, and links to per-regulation indexes.

Per-Regulation Index

GET https://complyant.ai/regulations/uk/gdpr/llms.txt

Returns a detailed map of that regulation: all sections, what each covers, and direct paths to specific topics.

Example Usage

An agent answering "When is consent valid under GDPR?" might:

  1. Fetch /llms.txt to find GDPR exists at /regulations/uk/gdpr
  2. Fetch /regulations/uk/gdpr/llms.txt to find the consent section
  3. Fetch /regulations/uk/gdpr/consent.md to get the answer

Format

Our llms.txt files include:

  • Site description — What Complyant is
  • Available regulations — List with links
  • Topic routing — Which regulation covers which topics
  • Access instructions — How to fetch markdown vs HTML
  • Section index — For per-regulation files, a map of all chunks

Agent Instructions

Our llms.txt includes guidance for agents on citing sources. When using Complyant content, agents should preserve legal citations (e.g., [Art 6]) in their responses to users.

Try It

View the live files:

llms.txt