Skip to content

Hadron Docs

Most AI tools forget everything when the session ends. Hadron is where they remember — and where you decide exactly who, and which agent, gets to see what.

Hadron holds what your agents and chatbots learn — who someone is, what was decided, what you told them last week — as knowledge you assemble in blocks and hand out deliberately, piece by piece. It outlives the conversation, it can be searched, and it is shared with the people and tools you choose and no others. Your team's AI knows your business without leaking it.

You do not need to install anything to start. If you can open a browser, you can build your first agent in the portal and watch it remember your name after a reload. The command line, the API and the MCP tools are all here too, further down and on the reference pages — for when you want them, not before.

Start where you are

  • I want to try it. Getting started — build a chatbot agent that remembers you, entirely in the portal. About twenty minutes. Nothing to install, though you will need an API key from an LLM provider such as Anthropic or OpenAI.
  • I want to know what this is first. What is Hadron? — what the platform does and who it is for.
  • My team already uses Hadron. Work with your team coordinator — talk to your team's AI coordinator in the portal. No setup at all.
  • I want my coding agent to use it. Add Hadron to Claude Code — one command and a browser sign-in. Other editors are in the how-to guides.
  • I need exact detail. Reference — the API, the CLI, the MCP tools, the data model.

How these docs are organised

These docs follow the Diátaxis framework, which splits documentation into four modes based on what you are trying to do:

  • Tutorials

    Learning-oriented lessons that take you step-by-step from zero to a working result.

  • How-to guides

    Task-oriented recipes for solving specific problems you already know you have.

  • Reference

    Information-oriented technical descriptions — APIs, schemas, configuration.

  • Explanation

    Understanding-oriented background on how and why Hadron works.

If you are the kind of reader who likes the shape of a system before its surface, entity architecture is the map.