Hermes Web UI

// Overview

Hermes Web UI is a self-hosted browser interface for Hermes Agent, the always-on autonomous agent that runs on your own server, keeps memory across sessions and is reachable from messaging apps. The project's goal is parity with the Hermes CLI: anything you can do from a terminal you can do from the browser, using your existing Hermes configuration, models and provider keys with no extra setup.

It is deliberately small in construction — Python on the server and vanilla JavaScript in the browser, with no framework, bundler or build step. The layout is three panels: sessions and navigation on the left, chat in the centre, and a workspace file browser with inline preview on the right. Model, profile and workspace controls sit in the composer footer so they stay visible while you type, a circular context ring shows token usage at a glance, and the remaining settings and session tools live in a Hermes Control Center launched from the sidebar.

Chat runs the Hermes agent in-process, reading your `HERMES_HOME` configuration directly rather than calling an external agent API — worth knowing when you already run a Hermes endpoint, because `HERMES_API_URL` is only read by the tasks/cron health probe. Two documented alternatives exist: add your endpoint as a custom OpenAI-compatible provider, or route browser chat through a running Hermes Gateway with `HERMES_WEBUI_CHAT_BACKEND=gateway`. Deployment options include a Nix flake and NixOS module, single- and multi-container Docker, and an SSH tunnel for secure remote access. The bootstrap supports Linux, macOS and WSL2; native Windows is not yet a supported bootstrap target.

// What it does

• Near-parity with the Hermes CLI from the browser, reusing your existing Hermes agent, models and provider configuration
• Three-panel layout: sessions sidebar, chat, and a workspace file browser with inline preview
• Composer-footer model, profile and workspace switches plus a circular context ring for token usage
• No build step — Python server and vanilla JS, so there is no bundler or framework to maintain
• `ctl.sh` daemon lifecycle for homelab and VM installs: start, status, logs, restart, stop, with a PID file and log under ~/.hermes
• Nix flake and NixOS module, plus single- and multi-container Docker deployments
• First-run onboarding wizard that walks through provider setup, with light and dark themes and a mobile-friendly layout

// Getting started

The repository bootstrap does the whole setup: it detects Hermes Agent (and offers to run the official installer if it is missing), prepares a Python environment, starts the server, waits for /health and drops you into an onboarding wizard.

git clone https://github.com/nesquena/hermes-webui.git hermes-webui
cd hermes-webui
python3 bootstrap.py

./start.sh

./ctl.sh start              # background daemon, PID at ~/.hermes/webui.pid
./ctl.sh status             # PID, uptime, bound host/port, log path, /health
./ctl.sh logs --lines 100   # tail ~/.hermes/webui.log
./ctl.sh restart
./ctl.sh stop

HERMES_WEBUI_HOST=0.0.0.0 ./ctl.sh start

hermes model

Commands and code are distilled from the project's own documentation — always check the official repo for the latest.

// When to use it

• Reach a server-resident Hermes agent from a laptop or phone browser instead of SSHing into a terminal
• Browse and preview the agent's workspace files next to the conversation that produced them
• Watch context usage while composing, and switch model or profile mid-session without restarting
• Run the UI as a managed daemon on a homelab box, with logs and health checks via ctl.sh

// How Hermes Web UI compares

Hermes Web UI alongside other open-source chat uis tools AI/TLDR tracks, ranked by GitHub stars.