Deploy and Host Hermes Agent on Railway

Hermes Agent is the open-source autonomous agent from NousResearch. This template deploys a full, self-hosted instance in one click — pinned to a known-good release and wrapped in a polished admin UI that supervises the agent gateway and the native Hermes dashboard behind a single login.

About Hosting Hermes Agent

Hosting Hermes Agent means running the agent gateway (hermes gateway) and the native React dashboard (hermes dashboard) as long-lived processes, fronted by an admin server that handles authentication, a setup wizard, live metrics, and file management. This template packages all of that into a single container: tini reaps zombies as PID 1, a Starlette admin app reverse-proxies the native dashboard, and everything the agent writes — config, keys, sessions, skills, memories, logs, cron, and workspace — is persisted to a /data volume. Providers, models, channel tokens, and keys are all configured through the web UI, so there are no config files to hand-edit.

Why Deploy Hermes Agent on Railway?

Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying Hermes Agent on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.

Hermes Agent benefits specifically from Railway's persistent volumes (its entire state lives on /data), one-click redeploys for version bumps, and built-in HTTPS domains for the admin UI and reverse-proxied dashboard.

Common Use Cases

• Running a personal, always-on AI agent reachable from messaging platforms with pairing approval.
• Self-hosting an agent with full control over providers, models, skills, and MCP servers.
• Prototyping and managing agent skills, cron jobs, and workspace files through a web UI instead of SSH.

Dependencies for Hermes Agent Hosting

• The upstream NousResearch/hermes-agent project (built into the image at a pinned HERMES_REF).
• A persistent volume mounted at /data for all agent state.
• An LLM provider key or OAuth (e.g. xAI / SuperGrok), configured after deploy through the setup wizard.

Deployment Dependencies

• NousResearch/hermes-agent — the agent this template deploys.
• Hermes releases — tags used for the HERMES_REF build/runtime variable.

Why These Dependencies?

Hermes Agent itself is the core application; the pinned release tag guarantees the pre-built dashboard and TUI assets in the image stay in sync with the agent code, so the first load is instant and in-dashboard updates behave predictably.

Getting Started After Deploy

1. Open the generated Railway domain.
2. Log in with username admin and the ADMIN_PASSWORD service variable.
3. Complete the setup wizard to add a provider, model, and any channels.

To update, bump the HERMES_REF variable to a newer tag from the Hermes releases page and redeploy. MIT licensed; Hermes Agent itself is MIT — see the upstream repository.