Install Hermes on macOS

What this page covers

This page gives a concrete macOS install path for Hermes Agent: install, reload the shell, choose a model provider, run a first chat, and verify the setup with hermes doctor before adding gateway, cron, memory, or voice features.

Hermes changes quickly, so treat the official documentation as the source of truth and use the built-in CLI to inspect the local install before editing config by hand.

Fast path

curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
source ~/.zshrc   # or source ~/.bashrc if that is your shell
hermes model
hermes --tui   # modern TUI
hermes         # classic CLI
hermes doctor
hermes config path
hermes config env-path

Practical checklist

  • Start with the desktop installer if the user wants the easiest macOS path; use the shell installer for command-line-only installs.
  • Run hermes model or hermes setup --portal immediately after install so provider and model details are written by Hermes, not guessed.
  • Keep secrets in ~/.hermes/.env; keep non-secret settings in ~/.hermes/config.yaml.
  • Use a small, verifiable first prompt such as asking Hermes to summarize the current repo or identify the main project file.
  • Only add gateway, cron, skills, memory, or voice after a normal chat works.

Common pitfalls

  • Skipping shell reload after install, leaving hermes off PATH.
  • Pasting API keys into chat or screenshots instead of using hermes auth / hermes config set.
  • Treating install success as workflow success; a working first chat is the real baseline.

Verification checklist

  • hermes --version returns a version.
  • hermes doctor reports no blocking setup errors.
  • A first hermes or hermes --tui conversation responds with the selected provider/model.
  • hermes config path and hermes config env-path point to the expected profile.

Official reference

https://hermes-agent.nousresearch.com/docs/getting-started/quickstart