« Back to Product

Documentation

AI Agents

The complete documentation is additionally available as Markdown optimized for AI agents and Large Language Models (LLMs), e.g. Claude, ChatGPT, Cursor or GitHub Copilot. The files are automatically regenerated on every update of the website and are therefore always up to date with this documentation.

In contrast to the HTML page or the PDF, the files contain only the content without any layout. Therefore, they require significantly fewer tokens and can be processed faster by AI agents.

Structure

The files follow the llms.txt standard:

File Description
llms.txt Index of all files with a short description and approximate size in tokens. The entry point for AI agents.
llms/function-index.md All functions (command reference and module functions) with signature and short description, each with a link to the file containing parameters and examples.
llms/modules/*.md One file per module, e.g. enocean.md or knx.md, including the module functions.
llms/functions/*.md The command reference, one file per group, e.g. management-variables.md.
llms-full.txt The complete documentation in a single file, e.g. for uploading into a knowledge base.

The file names are identical in all languages. The German version is available at /de/llms.txt.

Usage

Most AI agents can fetch the files directly. It is sufficient to tell the agent the address of the index, e.g. in the CLAUDE.md or AGENTS.md file of your module project:

The Symcon documentation for AI agents is available at https://www.symcon.de/en/llms.txt.
For PHP functions, load https://www.symcon.de/en/llms/function-index.md first.

Alternatively, the file llms-full.txt can be uploaded as a knowledge base into a project (e.g. Claude Projects or NotebookLM).

Files

Any questions?