Hermes Agent Quickstart

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

Overview

Hermes Agent is an AI agent that operates through a command-line interface, supporting multiple LLM providers and featuring tool access, messaging platform integration, and skill automation.

Installation

**Desktop (macOS/Windows):** Download from https://hermes-agent.nousresearch.com/desktop

**Command-line only:**

• Linux/macOS/WSL2/Android: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
• Windows PowerShell: iex (irm https://hermes-agent.nousresearch.com/install.ps1)

After installation, reload your shell: source ~/.bashrc or source ~/.zshrc

Provider Selection

Run hermes model for interactive setup. Key requirements: "a model with at least **64,000 tokens** of context."

Major supported providers include:

• Nous Portal (subscription, OAuth)
• OpenAI/Anthropic/Google
• OpenRouter, DeepSeek, xAI
• Ollama, LM Studio (local)
• AWS Bedrock, Azure Foundry
• Hugging Face, and 30+ others

Quick start: hermes setup --portal for Nous Portal with zero-config setup.

**Configuration storage:**

• Secrets -> ~/.hermes/.env
• Non-secret settings -> ~/.hermes/config.yaml

First Chat

hermes              # Classic CLI
hermes --tui        # Modern TUI interface

Verify: The agent responds with your chosen model and can use tools (terminal, file read, web search).

Core Commands

Key Features

**Slash commands:** Type / for autocomplete (/help, /tools, /model, /personality)

**Terminal access:** Agent can execute commands and show results

**Multi-line input:** Alt+Enter, Ctrl+J, or Shift+Enter

**Interruption:** New message during task execution switches focus; Ctrl+C also works

Advanced Setup (After Base Chat Works)

• **Messaging bots:** hermes gateway setup for Telegram, Discord, Slack, WhatsApp, Signal, Email
• **Sandboxing:** hermes config set terminal.backend docker
• **Voice:** Install extras, then /voice on with Ctrl+B to record
• **Skills:** hermes skills install openai/skills/k8s then /k8s [task]
• **MCP servers:** Configure via ~/.hermes/config.yaml

Troubleshooting

Common issues:

• Empty replies -> Run hermes model to verify provider auth
• Session won't resume -> Check hermes sessions list
• Gateway unresponsive -> Re-run hermes gateway setup

Recovery sequence: hermes doctor -> hermes model -> hermes setup -> hermes sessions list