Evey Setup
42-evey/evey-setup
Shell installer for a Hermes Agent stack with LiteLLM, Ollama, Qdrant and 29 community plugins
Evey Setup is a set of shell scripts that bootstraps a Hermes Agent stack with Docker, including a LiteLLM model proxy, Ollama, Qdrant, SearXNG and MQTT, plus 29 community plugins.
What Evey Setup does
The native flow runs bash setup.sh, which walks through API keys, scaffolds the directory, writes configs and clones hermes-agent. You then pick a service tier from three docker-compose templates and start the containers. bash install.sh runs all four phases at once, and flags such as --yes, --tier full and --plugins all support unattended installs. bash verify.sh checks the wiring and exits 0 on an overall pass.
The stack puts hermes-agent on port 8642, LiteLLM on 4000 for routing, fallbacks and budget limits, Ollama on 11434 for local GPU inference, Qdrant on 6333 for vector memory, SearXNG for private search, Mosquitto for MQTT events and ntfy for push notifications. The full tier adds n8n, Langfuse and Uptime Kuma. All ports bind to 127.0.0.1 only. The 29 community plugins are grouped into core, observability, social, memory, quality and extra categories.
Key features
- Phased scripts: setup, services, plugins and configure
- LiteLLM proxy with routing, fallbacks and budget limits
- Ollama local inference that falls back to CPU without an NVIDIA GPU
- Qdrant, SearXNG, MQTT and ntfy services for memory, search and events
- 29 community plugins in six categories
- verify.sh check after any install
When to use it
- Standing up a Hermes Agent with model routing and local inference on one machine
- Running an agent on free OpenRouter models behind a budget-limited proxy
- Scripting an unattended Hermes stack install for CI or a fresh server
Who it is for: Self-hosters who want a ready-made Docker stack around Hermes Agent instead of wiring each service by hand.
How it fits with Hermes Agent
Built for Hermes Agent: the installer clones hermes-agent and configures the surrounding services and plugins for it.
How to install Evey Setup
These commands are copied from the project's README. Check the repository for the latest steps before you run them.
git clone https://github.com/42-evey/evey-setup.git
cd evey-setup
bash setup.sh
bash verify.shRequirements: Docker 24.0+ with Compose v2, Git, 5 GB or more of free disk and an OpenRouter API key (the free tier works); an NVIDIA GPU is optional
FAQ
What is Evey Setup?
Evey Setup is a script-based installer that builds a complete Hermes Agent stack. It adds a model proxy, local inference, vector storage, search, messaging services and community plugins.
Does Evey Setup work with Hermes Agent?
Yes, it is built for hermes-agent and clones it during setup. The README calls the native hermes-agent flow the recommended path for all users.
What do I need to run Evey Setup?
You need Docker 24.0 or newer with Compose v2, Git, at least 5 GB of free disk and an OpenRouter API key, where the free tier works. An NVIDIA GPU is optional because Ollama falls back to CPU.
Similar deployment for Hermes Agent
All deploymentSkill that lets your AI coding agent self-host 2,200+ open-source apps, from provisioning to TLS
WolframRavenwolf Hermes Agent Home Assistant Add-onHome Assistant add-on that runs Hermes Agent with a dashboard, web terminal and messaging gateway
inceptionstack LowKey DeployerInstaller for self-hosted coding agents on AWS, including an experimental Hermes pack
Fractera FracteraSelf-hosted VPS installer and multi-agent coding environment with Hermes as its orchestrator
yuluyangguang1 Hermes PortablePortable Hermes Agent bundle for USB drives with bundled Python, Node.js, web UI and a config center
solomon2773 NoraSelf-hosted control plane for OpenClaw and Hermes agent fleets on Docker and Kubernetes
Related guides: How to run Hermes Agent securely · How to install Hermes Agent · Connect Hermes agents on several machines with Hermes Desktop