What is Hermes?

Hermes (by Nous Research) is an open-source AI agent that grows with you. Unlike a plain chatbot, it keeps persistent memory, builds reusable skills, and runs as a long-lived agent. In this kit it runs as a web dashboard — in Docker, on Ollama Cloud — that you open in the browser (no install, no GPU).

Three things make it different:

Autonomous — it can plan and carry out multi-step tasks with tools.
Persistent memory — it remembers across sessions, not just within one chat.
Self-improving skills — it can write down a procedure once and reuse it.

The files that make it tick

Hermes keeps its “brain” as plain files — which is exactly what makes it teachable. In this kit they live in the hermes repo, which you clone into VS Code (clone-projects → projects/hermes/). The running container is seeded from them, so you edit a file, push, redeploy, and the agent changes.

File / folder	What it is
`SOUL.md`	The agent’s persona and standing instructions — who it is.
`AGENTS.md`	Project-specific rules and conventions.
`config.yaml`	The Hermes configuration — provider and models.
`skills/`	Reusable `SKILL.md` procedures the agent can call.

(Memory and chat sessions are runtime — you see them on the dashboard’s Memory and Chat screens, not in the repo.)

Why it matters: the agent’s behaviour isn’t hidden in a cloud account — it’s a git repo you read, edit, and version. That’s the whole point of Track 1.

Beyond SOUL.md, a lot of what you want an agent to know is project-specific — that’s what AGENTS.md is for (project architecture, conventions, key paths). It’s a whole concept of its own → AGENTS.md.

Try it: open the Hermes dashboard at hermes.konomondo.org (log in trainee / hermes) and tour the Skills, Profiles, Memory and Tools screens. Then in VS Code run clone-projects and open projects/hermes/ — the editable versions of those files (SOUL.md, AGENTS.md, config.yaml, skills/).

Next: Skills — the reusable procedures that show up across every agent tool, including Claude Code.