By the end of this, you will have a personal AI agent running on your own machine or a cheap server. It will run tasks on a schedule, remember what it did last time, and you can talk to it from Telegram .

Before You Start : What You Need

• An API key from any LLM provider. Cheapest option to start: OpenRouter : sign up, add $5, and you get access to 200+ models including free ones. You can also use your existing OpenAI or Anthropic key.
• About 15 minutes for the initial setup.

Step 1 : Install Hermes Agent

Open your terminal (on Mac: press Cmd + Space, type Terminal). Paste this single command and hit Enter:

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

This installs everything automatically : Python, Node.js, and Hermes itself. It takes 2–3 minutes.

Once it finishes, run:

source ~/.bashrc

If you're on Mac using zsh (most Macs), run this instead:

source ~/.zshrc

✅ To confirm it worked, type hermes and press Enter. You should see a chat interface open.

Step 2 : Connect Your AI Model

Inside the Hermes terminal, type:

hermes model

This opens a menu. Pick your provider: