Hermes Agent — Setup Commands

1 — Install

Linux / macOS / WSL2

curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash

Reload shell after install

source ~/.bashrc

2 — Configure Model (Ollama)

Run model setup wizard

hermes model

Select Custom OpenAI-compatible endpoint
Base URL: http://localhost:11434/v1
Model: glm-4.7-flash:latest (or any Ollama model)

3 — Start Chatting

Launch CLI

hermes

4 — Telegram Integration

Approve pairing request from Telegram

hermes pairing approve telegram <PAIRING_CODE>

Start / check gateway

hermes gateway start
hermes gateway status

Create your bot at @BotFather on Telegram first, then paste the token during setup.

5 — Learning Loop Demo

Step 1 — Save something to memory

Remember that my name is Fahd and I prefer concise responses. Save this to your memory.

Step 2 — Start a fresh session

/new

Step 3 — Test recall

What do you know about me?

6 — Useful Commands

Re-run setup wizard

hermes setup

Change model anytime

hermes model

Check for issues

hermes doctor

Update to latest version

hermes update

Migrate from OpenClaw

hermes claw migrate

Resume last session

hermes --continue

⚠ Security Note

Hermes has full access to your terminal, files, and machine. Restrict Telegram access by setting allowed user IDs during setup, or run hermes setup gateway to configure it later. Do not run with open access on a production server.