The-Focus-AI

Hermes Body — Development skill for Claude Code

Development community

Reachy Mini front end for Nous Research's hermes-agent — OpenAI Realtime voice loop + robot motion tools, with hermes-agent as the knowledge backend.

How to install Hermes Body

This entry records only its repository, not the path inside it, so there is no exact command to give. Open The-Focus-AI/hermes-body and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Hermes Body does

Reachy Mini front end for Nous Research's hermes-agent — OpenAI Realtime voice loop + robot motion tools, with hermes-agent as the knowledge backend.

Alternatives in Development

Happy — Mobile and Web client for Claude Code, with realtime voice, encryption and fully featured 16.3k ★
Fidelity — Measure how faithfully a clone reproduces a site — pixel-diff plus motion-fidelity into one 0-100 score, a let 3.6k ★
Agentic Stack — One brain, many harnesses 2.2k ★

README

title: hermes-body emoji: 🧠 colorFrom: purple colorTo: blue sdk: static pinned: false short_description: Reachy Mini front end for Nous Research's hermes-agent tags:

reachy_mini
reachy_mini_python_app

hermes-body

A Reachy Mini front end for [Nous Research's hermes-agent](https://github.com/NousResearch/hermes-agent), modeled on [ClawBody](https://github.com/tomrikert/clawbody).

OpenAI Realtime is the embodied "brain" (sub-second voice loop, robot motion tools); hermes-agent is the "knowledge backend" reachable via a single `ask_hermes` tool. Cross-channel memory stays in sync via post-turn syncs.

See [PLAN.md](./PLAN.md) for the build plan and [reports/2026-05-03-hermes-body-reachy-mini-frontend.md](./reports/2026-05-03-hermes-body-reachy-mini-frontend.md) for the architectural rationale.

Install

Prereqs: [mise](https://mise.jdx.dev/getting-started/installation.html) and Docker Desktop. Mise will install Python 3.11 itself.

1. Local install (Mac, with sim)

git clone https://github.com/the-focus-ai/hermes-body
cd hermes-body
mise trust                  # one-time: allow this project's mise.toml
mise install                # provisions Python 3.11.x and the .venv
mise run install            # installs reachy-mini[mujoco] + hermes-body editable

The `.venv` is auto-activated by mise whenever you `cd` into the project. After this you can use either `mise run ` or call `hermes-body`/`python` directly.

2. Run hermes-agent in Docker

hermes-body talks to hermes-agent over its OpenAI-compatible HTTP API. State lives in a Docker named volume (`hermes_volume`) so config, memories, and skills survive container rebuilds.

2a. First-time setup (interactive wizard)

docker run -it --rm \
  -v hermes_volume:/opt/data \
  nousresearch/hermes-agent setup

This walks you through choosing an LLM provider (OpenRouter, DeepSeek, Anthropic, …), pasting that provider's API key, and selecting a default model. Setting