Use Docs with AI Tools
The HPC-AI.COM documentation site supports the llms.txt protocol. You can browse the docs directly or seamlessly integrate them into AI tools that support this protocol, allowing AI to provide more accurate answers based on HPC-AI.COM documentation.
1. Using Docs in Cursor
1.1 Configure the documentation data source
- Open your Cursor project and press
Command + L(macOS) orCtrl + L(Windows/Linux) to open the chat panel - Type
@in the input box and selectDocsfrom the popup menu - Click
Add new doc - Enter the documentation URL:
https://www.hpc-ai.com/doc/llms.txt - Click confirm and wait for the documentation to be indexed
1.2 Using the documentation
Once configured, type @Docs in the Cursor chat to select HPC-AI.COM documentation as context. You can ask AI questions about HPC-AI.COM APIs, model usage, and more — AI will answer based on the documentation content.
2. llms.txt Protocol Overview
2.1 Background
llms.txt is a web standard designed to help large language models access and understand website content more efficiently. By placing an llms.txt file in the root directory, a website can provide AI systems with clear navigation and content summaries, improving the accuracy of information retrieval.
2.2 File structure
The llms.txt file uses Markdown format and typically includes:
- Title: The name of the website or project
- Summary: A brief introduction to the website or project
- Section list: Important sections of the website, each with a link and brief description
Example from the HPC-AI.COM documentation site (full content at https://www.hpc-ai.com/doc/llms.txt):
# HPC-AI.COM Docs
> Complete documentation for the HPC-AI.COM platform. Quickstart, Model APIs, Fine-Tuning SDK, tutorials, FAQs, and more.
## Model APIs
### User Guides
- [Quickstart](https://www.hpc-ai.com/doc/docs/Model-APIs/User-Guides/QuickStart.md): ...
- [Text Models](https://www.hpc-ai.com/doc/docs/Model-APIs/User-Guides/Text-Models.md): ...
...
2.3 File purposes
/llms.txt
Provides an AI-friendly documentation navigation index, enabling LLMs to quickly discover and locate relevant content. It's essentially SEO for AI — users can obtain specific product information from HPC-AI.COM through general-purpose LLMs.
Individual .md files
Plain-text versions of each documentation page (e.g., https://www.hpc-ai.com/doc/docs/Model-APIs/User-Guides/QuickStart.md). AI tools can load these directly into context to get full documentation details.
2.4 Comparison with existing standards
llms.txt overlaps partially with robots.txt and sitemap.xml, but serves a different purpose:
- robots.txt: Instructs search engine crawlers which pages can or cannot be crawled; primarily focused on access control
- sitemap.xml: Provides a structural map of the website to help search engines understand page layout; primarily used for indexing
- llms.txt: Provides structured content overviews for large language models, helping AI systems better understand and process website information, improving AI interaction quality
3. Using with other tools
Other tools that support the llms.txt protocol can also use HPC-AI.COM documentation. For example, in ChatGPT, you can add https://www.hpc-ai.com/doc/llms.txt as a document link to your conversation context. In tools like Windsurf and Cline that support similar functionality, the process is the same — find the option to add external documentation or data sources and enter the llms.txt URL.
4. Further reading
- llms.txt protocol: https://llmstxt.org/
- Cursor @Docs: https://docs.cursor.com/context/@-symbols/@-docs