Getting Started with Hermes Agent
Install the self-improving, provider-agnostic AI agent that learns from your workflows.
Getting Started with Hermes Agent
Hermes Agent is an open-source autonomous agent built by Nous Research. It is designed to be the only assistant you need: it can chat in your terminal, send and receive messages on Telegram, Discord, Slack, WhatsApp, Signal, and Email, and it improves itself over time by saving reusable skills and building persistent memory of your preferences.
Hermes is model-agnostic, so you are not locked into a single provider. You can run it on a $5 VPS, a local desktop, a GPU server, or even Android via Termux. The same agent process can serve a personal chat interface, a team bot, and an always-on automation worker at the same time.
What Hermes Agent is
Hermes wraps several capabilities into one runtime:
- Conversational agent — multi-turn reasoning with tool use across files, web search, browsers, code execution, and messaging.
- Skill learner — after complex tasks, Hermes can create reusable skills that improve with each use. Skills follow the open agentskills.io standard.
- Memory system — cross-session recall via FTS5 session search, LLM summarization, and user modeling (Honcho dialectic integration).
- Messaging gateway — connect Telegram, Discord, Slack, WhatsApp, Signal, Email, and more to a single agent process.
- Automation scheduler — built-in cron for daily reports, nightly research, backups, and recurring tasks described in natural language.
- Subagent orchestration — spawn isolated workers for parallel pipelines, with Python scripts that call tools via RPC.
Strengths
- Provider freedom. Switch between 20+ providers — Nous Portal, OpenRouter, NVIDIA NIM, Kimi/Moonshot, MiniMax, Hugging Face, OpenAI, and custom endpoints — without changing code.
- Self-improving workflow. Skills created from real tasks get refined each time they run, so the agent gets cheaper and more reliable for repeated work.
- Terminal-first but not terminal-only. A polished TUI for power users plus messaging bridges for everyone else.
- Runs anywhere. Seven terminal backends: local, Docker, SSH, Singularity, Modal, Daytona, and Vercel Sandbox. Daytona and Modal offer serverless persistence that hibernates when idle.
- Research-ready tooling. Batch trajectory generation, RL environments, and trajectory compression for training the next generation of tool-calling models.
Requirements
Minimum practical setup:
- Git — the only hard prerequisite on non-Windows platforms.
- A supported OS: Linux, macOS, Windows, WSL2, or Android (Termux).
- A model provider or local endpoint:
- Easiest path: Nous Portal subscription.
- BYO-key path: OpenRouter, OpenAI, Anthropic, Kimi, MiniMax, etc.
- Local path: your own OpenAI-compatible endpoint or llama.cpp server.
- The installer auto-installs:
- Python 3.11 (via
uv) - Node.js 22
- ripgrep
- ffmpeg
- Python 3.11 (via
For Android/Termux, use the dedicated Termux guide because the install path is curated for mobile constraints.
How to get started
1. Install Hermes Agent
Recommended: download the Hermes Desktop installer for macOS or Windows to get both the GUI and CLI in one step.
For command-line only install:
Linux / macOS / WSL2 / Android (Termux):
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
Windows (native PowerShell):
iex (irm https://hermes-agent.nousresearch.com/install.ps1)
Then reload your shell:
source ~/.bashrc # or source ~/.zshrc
2. Pick your provider
The fastest one-command setup uses Nous Portal:
hermes setup --portal
That logs you in, sets Nous as your provider, and turns on the Tool Gateway (web search, image generation, TTS, cloud browser) in one step.
If you already know your provider, run:
hermes model
3. Start chatting
hermes
Run a real task — ask it to explain a file, write a script, or search the web — and verify it responds before layering on more features.
4. Connect channels (optional)
Once the base chat works, set up the messaging gateway:
hermes gateway setup
Then connect Telegram, Discord, Slack, or WhatsApp so you can talk to the same agent from anywhere.
5. Enable automation (optional)
Use the built-in scheduler for recurring tasks:
hermes cron add "every day at 8am" "Summarize my unread newsletters"
Resources
Official
- Hermes Agent website
- Official documentation
- Quickstart guide
- Installation guide
- Termux / Android guide
- Learning path
- GitHub repository
- Hermes Desktop download
- Nous Portal
Video and community
- Hermes Agent Tutorials & Use Cases YouTube playlist
- Onchain AI Garage Masterclass (installation walkthrough)
- Nous Research Discord
Reference and mirrors
- Hermes Agent docs mirror / cheatsheet (community-maintained)
- mudrii/hermes-agent-docs (comprehensive community docs)