Hermes Agent: The Open-Source Autonomous AI Agent
The AI agent framework that grows with you.
Install it on your server, give it your messaging accounts, and it becomes a persistent personal agent — learning your projects, building its own skills, and reaching you wherever you are. Not a chatbot. Not a copilot. An agent that lives on your machine and gets smarter every day.
Core Capabilities for Full AI Agent Workflows
Hermes Agent combines persistent memory, automated skill creation, and multi-platform reach into a single open-source package.
Lives Where You Do
Telegram, Discord, Slack, WhatsApp, Signal, Email, CLI — and a growing list of platforms. Start on one, pick up on another.
Grows the Longer It Runs
Persistent memory and auto-generated skills — it learns your projects and never forgets how it solved a problem.
Scheduled Automations
Natural language cron scheduling for reports, backups, and briefings — running unattended through the gateway.
Delegates & Parallelizes
Isolated subagents with their own conversations, terminals, and Python RPC scripts for zero-context-cost pipelines.
Real Sandboxing
Five backends — local, Docker, SSH, Singularity, Modal — with container hardening and namespace isolation.
Full Web & Browser Control
Web search, browser automation, vision, image generation, text-to-speech, and multi-model reasoning.
Autonomous AI Agent Capabilities for Real-World Workflows
Go beyond basic chat. Hermes Agent is built for real-world execution, automation, and integrations.
Core Foundations
Persistent memory, safe sandboxing, and context discovery
Persistent Memory System
Remembers your preferences, project details, and historical context across sessions (MEMORY.md / USER.md).
Smart Context Understanding
Automatically loads project context files and supports @ references for files, folders, git diffs, and web URLs.
Modular Tools & Skills
Rich built-in toolsets combined with on-demand knowledge skills, flexibly enabled as needed.
Security Checkpoints
Automatically snapshots your workspace directory before making changes, supporting one-click rollback.
Automation & Workflows
Background tasks, scheduling, and multi-agent systems
Scheduled Tasks (Cron)
Schedule workflows via natural language or cron expressions for fully automated executions.
Subagent Delegation
Concurrently execute multiple child agent instances to run parallel workstreams.
Sandboxed Code Execution
Run Python scripts inside secure, isolated sandboxes to complete complex multi-step actions.
Batch Processing
Process massive prompts in parallel, optimized for evaluation and dataset generation.
Web & Multimedia
Speech, vision, image generation, and browser control
Voice Interactions
Supports microphone input and voice responses over terminal, Discord channels, and messaging apps.
Browser Automation
Complete browser control (local/cloud Chrome) to navigate pages, fill forms, and extract data.
Vision Capabilities
Supports clipboard image pasting to let the agent analyze, debug, or describe screenshots.
Image Generation
Integrated top-tier models (FLUX, Ideogram, Recraft) via FAL.ai for rapid asset creation.
Speech Synthesis & TTS
Multi-engine speech synthesis, text-to-speech, and automatic voice message transcription.
Integrations & Customization
MCP servers, IDE plugins, provider routing, and custom personalities
Broad Integrations
MCP server protocol, external memory backends, IDEs (VS Code/Zed), and OpenAI-compatible API endpoint.
Intelligent Routing
Dynamic provider routing, automatic failovers, credential pools, and Claude prompt caching.
Personality & Skins
Fully customizable agent personality (SOUL.md), custom CLI themes/skins, and spinner styles.
Extensible Plugins
Seamlessly add custom tools, lifecycle hooks, and integrations via plugins without core modifications.
An Ever-Expanding Universe of Skills
Discover, search, and install from 92,039 skills across 12 registries
Registries / Providers
Manage Apple Notes via memo CLI: create, search, edit.
Drive the macOS desktop in the background — screenshots, mouse, keyboard, scroll, drag.
Delegate coding to Claude Code CLI (features, PRs).
Infographics: 21 layouts x 21 styles.
Control a running TouchDesigner instance via twozero MCP — 36 native tools.
Read, search, create, and edit notes in the Obsidian vault.
Search arXiv papers by keyword, author, category, or ID.
Query Polymarket: markets, prices, orderbooks, history.
Fetch weather forecasts and compare multiple sources for market predictions.
Scan emails, generate inbox summaries, and post structured briefings to Slack daily.
Automate HTML generation, package uploads, and SSH into VPS for instant web deployment.
Intercept agent execution and ask for user approval before invoking tools.
What People Are Actually Building
Real stories from the Hermes Agent community, scraped from X, Discord, GitHub, and Substack.
Every weekday at 9am, summarize my inbox and post to Slack
“An agent that grows with you — not marketing fluff; it literally writes markdown skill files when it solves hard problems. Natural-language cron: 'every weekday at 9am, summarize my inbox and post to Slack.'”
Told it to Google me and ship a landing page to my VPS
“told it to Google me and then build a landing page based on what it found and that was genuinely mind blowing because it ran the searches, found links, created the page, SSH'd into my VPS, uploaded the page, then texted me when it was done. what?!”
$100 → $216 in 48h with a self-learning weather bot
“I turned $100 into $216 in less than 48 hours with a self-learning weather trading bot. Hermes scans weather markets every 60 mins, compares 3 forecast sources per location, buys undervalued temperature buckets and flips for profit.”
Built converse mode so my agent thinks before it acts
“My Hermes agent used to start executing the moment I hit enter... description of a vague idea and it would immediately write files. So I built converse mode. One plugin, two commands. The agent won't touch a single tool until you say so.”
Get Started in Minutes
No prerequisites. Works on Linux, macOS & WSL2. Installs everything automatically.
Install
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bashInstalls uv, Python 3.11, clones the repo, sets up everything. No sudo needed.
Configure
hermes setuphermes modelConnect to Nous Portal (OAuth), OpenRouter (API key), or your own endpoint. Interactive setup wizard.
Start Chatting
hermesFull interactive CLI with tools, memory, and skills. That's it.
Go Multi-Platform (Optional)
hermes gateway setuphermes gatewayhermes gateway installWalk through connecting Telegram, Discord, Slack, or WhatsApp. Runs as a systemd service.
Keep It Up to Date
hermes updatePulls the latest changes and reinstalls dependencies. Run anytime to get new features and fixes.
⚠️ Native Windows support is experimental. Please install WSL2 and run Hermes Agent from there.
Ready to Run Your Own Agent?
Self-hosted, open source, and free forever. One command to install, one command to start.
View on GitHub