Hermes Agent Quickstart — From Zero to a Working Setup

Hermes Agent Quickstart — From Zero to a Working Setup

Agent Daily · Tutorial · June 2026
Hermes Agent Quickstart Guide hero image
📖 Tutorial / Guide 🏢 Nous Research 🏷 Hermes Agent · Installation · AI Providers · Skills · Gateway

1 What Is Hermes Agent?

Hermes Agent is a fully open-source AI assistant built by Nous Research that lives in your terminal, on your messaging platforms, and inside your workflows. Unlike cloud-only AI chatbots, Hermes runs locally and has direct access to your file system, terminal, browser, and a rich ecosystem of tools and skills.

Think of it as an AI-native operating layer: you install it once, connect your preferred LLM provider (30+ supported), and get an agent that can read files, run shell commands, manage Git workflows, search the web, generate images, control smart home devices, and much more — all from a single conversational interface.

💻
Your Machine
CLI / TUI
🤖
Hermes Agent
Tools & Skills
🧠
LLM Provider
30+ Options
💡 Key concept: Hermes is provider-agnostic. You can switch between OpenAI, Anthropic, Google, DeepSeek, local models, or any OpenAI-compatible endpoint at any time with hermes model — zero lock-in.

2 Who This Is For

This quickstart guide is designed for anyone who wants the shortest path from zero to a working Hermes setup. Specifically:

  • Brand new users — you've heard about Hermes and want to get it running fast
  • Provider switchers — you already have an API key and don't want to waste time on config mistakes
  • Team/bot operators — you're setting up Hermes as a shared assistant or always-on workflow
  • Frustrated installers — you've been in the "it installed, but it does nothing" limbo before
Rule of thumb: If Hermes cannot complete a normal chat, do not add more features yet. Get one clean conversation working first, then layer on gateway, cron, skills, voice, or routing.

3 The Fastest Path

Pick the row that matches your goal and follow the steps. Don't overthink it — you can always reconfigure later.

Goal Do This First Then Do This
Just get Hermes working hermes setup Run a real chat and verify it responds
I already know my provider hermes model Save the config, then start chatting
I want a bot / always-on hermes gateway setup after CLI works Connect Telegram, Discord, Slack, etc.
I want a local model hermes model → custom endpoint Verify endpoint, model name, and context length
Multi-provider fallback hermes model first Add routing/fallback only after base chat works

4 Install Hermes Agent

Option A: Hermes Desktop (macOS / Windows — Recommended)

The easiest path. Download the Hermes Desktop installer from the official website and run it. This installs both the command-line and desktop applications.

Option B: Command-Line Only

For Linux, macOS, WSL2, or Android (Termux):

curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash

For Windows (native PowerShell):

iex (irm https://hermes-agent.nousresearch.com/install.ps1)

After installation finishes, reload your shell:

source ~/.bashrc # or source ~/.zshrc
⚠️ Android / Termux users: See the dedicated Termux guide for the tested manual installation path, supported extras, and Android-specific limitations.
💡 For detailed installation options, prerequisites, and troubleshooting, see the full Installation guide.

5 Choose a Provider

This is the single most important setup step. Use the interactive model selector:

hermes model

Easiest Path: Nous Portal

One subscription covers 300+ models plus the Tool Gateway (web search, image generation, TTS, cloud browser). On a fresh install:

hermes setup --portal

That logs you in via OAuth, sets Nous as your provider, and turns on the Tool Gateway — all in one command.

Setup Modes

On a fresh install, hermes setup offers three modes:

  • Quick Setup (Nous Portal) — Free OAuth login, no API keys needed. Sets up a model plus the Tool Gateway tools. The recommended fast path.
  • Full Setup — Walk through every provider, tool, and option yourself (bring your own keys).
  • Blank Slate — Everything starts disabled except the bare minimum: provider & model, File Operations, and Terminal toolset. Pick this when you want a minimal, fully-controlled agent and intend to enable only what you need.

Popular Providers at a Glance

Provider What It Is How to Set Up
Nous Portal Subscription-based, zero-config OAuth login via hermes model
Anthropic Claude models directly OAuth (Max plan) or API key
OpenAI Codex ChatGPT OAuth, Codex models Device code auth via hermes model
OpenRouter Multi-provider routing Enter your API key
Google AI Studio Gemini models Set GOOGLE_API_KEY
DeepSeek Direct DeepSeek API Set DEEPSEEK_API_KEY
xAI Grok models Set XAI_API_KEY or OAuth
GitHub Copilot Copilot subscription OAuth via hermes model
Custom Endpoint vLLM, Ollama, SGLang, etc. Set base URL + API key

Hermes supports 30+ providers in total, including Hugging Face, AWS Bedrock, Azure Foundry, Alibaba Cloud, Kimi/Moonshot, LM Studio, NVIDIA NIM, and many more. See the full Providers page for the complete catalog.

⚠️ Minimum context: 64K tokens. Hermes requires a model with at least 64,000 tokens of context. Models with smaller windows will be rejected at startup. If running a local model, set --ctx-size 65536 (llama.cpp) or -c 65536 (Ollama).

How Settings Are Stored

Hermes separates secrets from normal config:

  • Secrets and tokens~/.hermes/.env
  • Non-secret settings~/.hermes/config.yaml
# The easiest way to set values correctly hermes config set model anthropic/claude-opus-4.6 hermes config set terminal.backend docker hermes config set OPENROUTER_API_KEY sk-or-...
Tip: You can switch providers at any time with hermes model — no lock-in. The right value goes to the right file automatically.

6 Run Your First Chat

hermes # classic CLI hermes --tui # modern TUI (recommended)

You'll see a welcome banner with your model, available tools, and skills. Try a prompt that's specific and easy to verify:

# Good first prompts to try: Summarize this repo in 5 bullets and tell me what the main entrypoint is. Check my current directory and tell me what looks like the main project file. Help me set up a clean GitHub PR workflow for this codebase.

What Success Looks Like

  • ✅ The banner shows your chosen model/provider
  • ✅ Hermes replies without error
  • ✅ It can use a tool if needed (terminal, file read, web search)
  • ✅ The conversation continues normally for more than one turn
If that works, you're past the hardest part. Everything from here on is layering capabilities onto a working foundation.
💡 Pick your interface: Hermes ships with two terminal interfaces — the classic prompt_toolkit CLI and a newer TUI with modal overlays, mouse selection, and non-blocking input. Both share the same sessions, slash commands, and config. Try each with hermes vs hermes --tui.

7 Verify Sessions Work

Before moving on, make sure session resume works:

hermes --continue # Resume the most recent session hermes -c # Short form

This should bring you back to the session you just had. If it doesn't, check whether you're in the same profile and whether the session actually saved. This matters later when you're juggling multiple setups or machines.

Tip: Use hermes sessions list to see all saved sessions and confirm your work is persisting.

8 Key Features to Try

Terminal Access

The agent can run shell commands on your behalf:

❯ What's my disk usage? Show the top 5 largest directories.

Hermes runs the commands and shows results inline — no copy-pasting between windows.

Slash Commands

Type / to see an autocomplete dropdown of all commands:

CommandWhat It Does
/helpShow all available commands
/toolsList available tools
/modelSwitch models interactively
/personality pirateTry a fun personality
/saveSave the conversation

Multi-Line Input

Press Alt+Enter, Ctrl+J, or Shift+Enter to add a new line. Shift+Enter requires a terminal that sends it as a distinct sequence (Kitty, foot, WezTerm, Ghostty by default; iTerm2 and Alacritty need Kitty keyboard protocol enabled). Alt+Enter and Ctrl+J work everywhere.

Interrupt the Agent

If the agent is taking too long, type a new message and press Enter — it interrupts the current task and switches to your new instructions. Ctrl+C also works.

9 Add the Next Layer

Only do this after the base chat works. Pick what you need:

🤖 Messaging Gateway (Bot / Shared Assistant)

hermes gateway setup # Interactive platform configuration

Connect to Telegram, Discord, Slack, WhatsApp, Signal, Email, Home Assistant, or Microsoft Teams.

🔧 Tools & Automation

  • hermes tools — tune tool access per platform
  • hermes skills — browse and install reusable workflows
  • Cron — only after your bot or CLI setup is stable

🛡️ Sandboxed Terminal

For safety, run the agent's terminal inside Docker or on a remote server:

hermes config set terminal.backend docker # Docker isolation hermes config set terminal.backend ssh # Remote server

🎙️ Voice Mode

# Install voice extras cd ~/.hermes/hermes-agent uv pip install -e ".[voice]" # Then in the CLI: /voice on, press Ctrl+B to record

Includes faster-whisper for free local speech-to-text.

📦 Skills System

Skills are on-demand instruction documents that teach Hermes how to do specific tasks — deploy to Kubernetes, open a GitHub PR, fine-tune a model, search for GIFs. Each is a SKILL.md file with a name, description, and step-by-step procedure.

hermes skills browse # List everything available hermes skills search kubernetes # Find by keyword hermes skills install openai/skills/k8s # Install one (security scan first)

Every installed skill becomes a slash command automatically:

/k8s deploy the staging manifest # Run the skill with a request /k8s # Load it and let Hermes ask what you need

🔌 MCP Servers

Add to ~/.hermes/config.yaml:

mcp_servers: github: command: npx args: ["-y", "@modelcontextprotocol/server-github"] env: GITHUB_PERSONAL_ACCESS_TOKEN: "ghp_xxx"

✏️ Editor Integration (ACP)

ACP support ships with the standard [all] extras, so the curl installer already includes it:

hermes acp

10 Common Failure Modes & Recovery

These are the problems that waste the most time. Know them, avoid them.

Symptom Likely Cause Fix
Empty or broken replies Provider auth or model wrong Run hermes model again
Custom endpoint returns garbage Wrong URL, model name, or incompatible API Verify endpoint in a separate client first
Gateway starts but no messages Bot token, allowlist, or platform setup incomplete Re-run hermes gateway setup
Can't find old session Switched profiles or session didn't save Check hermes sessions list
Model unavailable or odd fallback Routing/fallback too aggressive Keep routing off until base is stable
hermes doctor flags issues Config values missing or stale Fix config, retest plain chat

Recovery Toolkit

When something feels off, run these commands in order:

hermes doctor # Diagnose issues hermes model # Reconfigure provider/model hermes setup # Full setup wizard hermes sessions list # Check session history hermes --continue # Resume last session hermes gateway status # Check bot gateway
✅ That sequence gets you from "broken vibes" back to a known state fast.

11 Quick Reference

CommandDescription
hermesStart chatting
hermes modelChoose your LLM provider and model
hermes toolsConfigure which tools are enabled per platform
hermes setupFull setup wizard (configures everything at once)
hermes doctorDiagnose issues
hermes updateUpdate to latest version
hermes gatewayStart the messaging gateway
hermes --continueResume last session
hermes skillsBrowse and install reusable workflows
hermes acpLaunch editor integration (ACP)

🎯 Key Takeaways

🔑 Key Takeaways

  • Install in one command — a single curl (or PowerShell iex) gets you from nothing to a working install in under a minute.
  • Nous Portal is the fastest pathhermes setup --portal handles OAuth login, provider setup, and Tool Gateway in one shot.
  • 30+ providers, zero lock-in — switch between Anthropic, OpenAI, Google, DeepSeek, Hugging Face, local models, and many more with hermes model.
  • 64K minimum context is mandatory — Hermes requires at least 64,000 tokens of context for multi-step tool-calling workflows.
  • Get one clean chat working first — don't add gateway, cron, skills, or routing until the base CLI chat responds without errors.
  • Two interfaces available — the classic CLI (hermes) and the modern TUI (hermes --tui) share the same sessions, commands, and config.
  • Skills are first-class — browse, install, and use skills as slash commands. The agent also auto-loads matching skills during conversation.
  • Messaging gateway supports 7+ platforms — Telegram, Discord, Slack, WhatsApp, Signal, Email, Home Assistant, and Teams are all supported.
  • Sandboxing is built in — run the agent's terminal in Docker or on a remote SSH server for safety.
  • hermes doctor is your best friend — when something feels off, it diagnoses config issues and suggests fixes.

🔗 Resources & Links