hermes-agent-helm: deployment tool for Hermes Agent

Helm chart that runs Hermes Agent on Kubernetes, with agent teams and Copilot or Codex sign-in

What hermes-agent-helm does

The chart installs from an OCI registry (ghcr.io/jyje/hermes-agent-helm) or from a Helm repository published to GitHub Pages. Provider settings live in values.yaml, with examples for OpenAI-compatible endpoints, Anthropic, Gemini, OpenRouter, NVIDIA and DeepSeek. A Discord or Telegram bot token starts Hermes' outbound gateway with a Kubernetes-managed lifecycle, and an optional device-login bootstrap for GitHub Copilot and OpenAI Codex sends a one-time link and code to the Discord home channel or the logs, then stores refreshable credentials in HERMES_HOME.

Defaults suit small clusters: one replica, modest requests and a small PVC. Because Hermes is a single-instance personal agent, the chart scales up rather than out, and several instances can be grouped into a Hermes Team that shares context over a common gateway channel. The Hermes web dashboard can be published through an Ingress or HTTPRoute behind a password, Nous Portal OAuth or OpenID Connect. CI installs the chart on a kind cluster and runs a test Job that calls hermes doctor.

Key features

Single helm install from an OCI registry or a Helm repository
Provider configuration through values.yaml, with no provider baked in
Discord or Telegram gateway with a Kubernetes-managed lifecycle
Optional device-login bootstrap for GitHub Copilot and OpenAI Codex
Hermes Team: several instances sharing context over a gateway channel
Optional dashboard exposure through Ingress or HTTPRoute with sign-in

When to use it

Running Hermes Agent on a homelab or single-node Kubernetes cluster
Signing in with a Copilot or Codex account instead of an API key
Running a team of Hermes instances on one cluster
Publishing the Hermes web dashboard behind OpenID Connect

Who it is for: Kubernetes users who want to deploy Hermes Agent with Helm and manage it as a normal cluster workload.

How it fits with Hermes Agent

It is a deployment chart built specifically for Hermes Agent, covering its gateway, dashboard, credentials directory and multi-instance teams.

How to install hermes-agent-helm

These commands are copied from the project's README. Check the repository for the latest steps before you run them.

helm install hermes-agent oci://ghcr.io/jyje/hermes-agent-helm/hermes-agent \
  --namespace hermes-agent --create-namespace \
  --set-string env.OPENAI_API_KEY='sk-...' \
  --wait

Requirements: Helm and a Kubernetes cluster; the quick start sets a provider key such as env.OPENAI_API_KEY at install time

Note: The README states this is a community-powered chart, not an official Nous Research release.

FAQ

What is hermes-agent-helm?

hermes-agent-helm is a Helm chart for running Hermes Agent on Kubernetes. It exposes provider, gateway and dashboard settings through values.yaml.

Does hermes-agent-helm work with Hermes Agent?

Yes. It is built specifically for Hermes Agent. The README states it is a community-powered chart and not an official Nous Research release.

How do I install hermes-agent-helm?

Run helm install against the OCI registry path with a namespace and a provider key, as shown in the install command. You can also add the chart's GitHub Pages Helm repository and install by name.