Hermes Agent Setup Guide

What is Hermes Agent?

Hermes Agent by Nous Research is the only AI agent with a built-in learning loop. Unlike static agents, Hermes creates skills from experience, improves them during use, and maintains its own skill library through autonomous curation.

Prerequisites

• ✓
  Python 3.10+ - Python Setup Guide
• ✓
  Node.js 18+ - Node.js Setup Guide
• ✓
  Git - GitHub Setup Guide
• ✓
  LLM API Key - Anthropic, OpenAI, or local Ollama
• ✓
  Virtual Environment - Virtual Environments Guide

Step 1: Install Hermes Agent

Clone the official repository and install dependencies:

# Clone the repository
git clone https://github.com/NousResearch/Hermes-Agent.git
cd Hermes-Agent

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Step 2: Configure Hermes

Create your configuration file:

# Copy example config
cp config.example.yaml config.yaml

# Edit configuration
nano config.yaml

Key settings to configure:

# LLM Configuration
llm:
  provider: anthropic  # or openai, ollama, bedrock
  model: claude-3-7-sonnet-20260101
  api_key: ${ANTHROPIC_API_KEY}

# Terminal Backend (where code executes)
terminal:
  backend: docker  # or local, ssh, modal, daytona, vercel

# Memory & Learning
memory:
  enabled: true
  persistence: sqlite  # or postgres
  curator_cycle_days: 7

# Messaging Transport
transport:
  platform: telegram  # or discord, slack, whatsapp, signal
  token: ${TELEGRAM_BOT_TOKEN}

Step 3: Install Skills

Hermes comes with base skills. Add community skills for more capabilities:

# Install wondelai skills (380+ skills, actively maintained)
git clone https://github.com/wondelai/skills.git ./custom_skills

# Or install litprog-skill for literate programming
git clone https://github.com/NousResearch/hermes-agent/tree/main/skills.git ./custom_skills/litprog

Popular skill categories:

• • Code Generation: Python, JavaScript, TypeScript, Rust
• • Web Automation: Browser control, API calls, scraping
• • Data Analysis: CSV/JSON parsing, charts, statistics
• • File Operations: Read/write, organize, convert formats
• • Communication: Email, Slack, Discord messaging

Step 4: Connect Messaging Platform

Choose your primary interface (18 platforms supported):

Step 5: Start Hermes Agent

Launch your self-improving agent:

# Start Hermes
python -m hermes.agent

# Or run with specific profile
python -m hermes.agent --profile production

# Enable verbose logging for debugging
python -m hermes.agent --log-level debug

First run checklist:

• ✓ Agent connects to LLM successfully
• ✓ Messaging platform connected
• ✓ Terminal backend initialized (Docker/local)
• ✓ Memory database created
• ✓ Skills loaded from registry

Step 6: Test Your Agent

Send your first message to Hermes:

Troubleshooting

📚 More Resources

Official Links