When deploying Hermes Agent with Docker, the focus is not on "container bootability", but on persistence of configuration directories, working directories, file mounts, and environment variables. Many Docker issues in the community are stuck: the host file is not visible in the container, or the Telegram/session configuration is lost after restarting.
Deployment points for novices
- Persist the volume of the
~/.hermesor equivalent configuration directory. - Explicitly hang the item directory that Hermes wants to read and write to the container.
- API key, Telegram token, allowed users into environment variables or env files.
- If you run a gateway, you can check the container restart policy and logs.
When is Docker not recommended
If you're just starting out with Hermes Agent and just want to try it out in the native CLI, it's easier to install it directly. Docker is better suited for VPS, long-term gateways, team bots, or scenarios where isolation permissions are required.
In a word: Docker deploys Hermes Agent, the core is volume and env, not just running images.
Official open source address: https://github.com/NousResearch/hermes-agent; Official document entrance: https://hermes-agent.nousresearch.com/.
