Multi-model chat
Claude, GPT, Gemini, Codex, and local models in one conversation flow.
Open-source UI for Hermes Agent
Your AI agent's command center
Chat, memory, skills, terminal, and files — one interface for everything your agent does.
Setup
Get running in under 3 minutes
Pick your backend, install the workspace, and you're live.
Step 1 — Start your backend
Full features — sessions, memory, skills, config, and streaming. Use our fork for extended gateway endpoints.
terminal
$git clone https://github.com/outsourc-e/hermes-agent.git
$cd hermes-agent
$pip install -e .
$hermes setup# pick your provider
$hermes --gateway# starts on :8642
Upstream NousResearch/hermes-agent also works — basic chat without sessions/memory/skills.
Step 2 — Install Hermes Workspace
terminal
$git clone https://github.com/outsourc-e/hermes-workspace.git
$cd hermes-workspace
$pnpm install
$pnpm dev# opens on localhost:3000
Screenshots
The interface, section by section
Real product screens across the core workflow.
Chat
Multi-model conversations with real-time tool activity.
Memory
Browse and search what your agent remembers.
Terminal
Shell access inside the workspace.
Skills
Browse and launch 100+ agent skills.
Features
Built for agents that do real work
A single interface for the parts of Hermes Agent that usually get split across terminals, tabs, and logs.
6
core workflow surfaces
100+
skills to browse
PWA
mobile parity
OSS
MIT licensed
Memory & skills browser
Persistent memory plus a deep catalog of 100+ skills your agent can pull from.
Integrated terminal
Run commands where your agent lives without leaving the workspace.
Real-time tool cards
Watch your agent think and act with live tool states streaming into the UI.
Mobile-first PWA
Full feature parity on desktop, tablet, or phone with an installable web app.
Works with any backend
OpenAI-compatible APIs, Hermes gateway, or portable mode with your own routing.