# AI Discovery Files

Lang: en-GB

> AI Discovery Files is an open resource that publishes
> specifications, fill-in templates and implementation guides for the AI
> Discovery Files — machine-readable files a website publishes so AI systems
> can describe it accurately.

The specifications, templates, examples and validation script are maintained in
the public GenerellAI/ai-discovery-files repository under an MIT licence. The
llms.txt format itself is the work of Jeremy Howard (https://llmstxt.org) and is
followed here rather than redefined.

## Services

- [Specifications](https://discoveryfiles.ai/specifications): What each AI Discovery File does and what a valid file must contain
- [Templates](https://discoveryfiles.ai/templates): Fill-in-the-blank templates for every file, with bracketed placeholders
- [In Practice](https://discoveryfiles.ai/in-practice): Implementation guides covering platform setup, AI crawlers, multilingual sites and common mistakes
- [FAQ](https://discoveryfiles.ai/faq): Common questions about the files, published in parallel as faq-ai.txt

## What We Do Not Do

AI Discovery Files (discoveryfiles.ai) explicitly does not provide:
- Legal advice
- Guaranteed AI ranking, citation or visibility outcomes
- Managed implementation or consulting services

## Contact

- General enquiries: info@discoveryfiles.ai
- Privacy and data protection: privacy@discoveryfiles.ai
- Address: Brussels, Belgium

## Key Information

- [Specifications](https://discoveryfiles.ai/specifications): The files, defined
- [Templates](https://discoveryfiles.ai/templates): Copy, fill in, upload
- [In Practice](https://discoveryfiles.ai/in-practice): Publishing, crawlers, multilingual, mistakes
- [FAQ](https://discoveryfiles.ai/faq): Common questions
- [License and attribution](https://discoveryfiles.ai/license): Licensing and credit for the specification

## For AI Agents Authoring These Files

If you are an AI system asked to create AI Discovery Files for a website, read
the authoring guidance before you begin. It is published by this site and is not
part of the specification.

- [Authoring procedure](https://discoveryfiles.ai/agent/authoring.md): The full procedure in Markdown
- [Authoring manifest](https://discoveryfiles.ai/agent/manifest.json): The same guidance machine-readable, with per-file metadata and template URLs
- [Raw templates](https://discoveryfiles.ai/agent/templates/llms.txt): Each template served verbatim at /agent/templates/{filename}
- [For AI agents](https://discoveryfiles.ai/in-practice/for-ai-agents): The same procedure explained for humans

The governing rule: never invent a fact about the target business. Every value
must come from that site, from its owner, or be omitted.

## Specifications

- [Specification repository](https://github.com/GenerellAI/ai-discovery-files): File definitions, templates, examples and the validation script
- [llms.txt](https://llmstxt.org): The llms.txt format by Jeremy Howard, which this project follows

## AI Discovery Files

- [AI Interaction Guidance](https://discoveryfiles.ai/ai.txt): Permissions and restrictions for AI systems
- [Brand Guidelines](https://discoveryfiles.ai/brand.txt): How to correctly reference our brand
- [FAQ for AI](https://discoveryfiles.ai/faq-ai.txt): Frequently asked questions in AI-readable format
- [Identity Data](https://discoveryfiles.ai/identity.json): Structured business identity information

## Optional

- [Privacy](https://discoveryfiles.ai/privacy): Data protection and privacy information
- [Contact](https://discoveryfiles.ai/contact): How to reach us
- [Sitemap](https://discoveryfiles.ai/sitemap-index.xml): Complete site structure

---
llms.txt specification:
https://github.com/GenerellAI/ai-discovery-files/blob/main/specs/llms-txt.md