An Agent That Grows With You
Hermes Agent is not a coding copilot tethered to an IDE or a chatbot wrapper. Hermes Agent is an autonomous AI agent that lives on your server, remembers what Hermes Agent learns, and gets more capable the longer you use Hermes Agent.
Core Features
Hermes Agent is more than a chatbot — Hermes Agent is a complete autonomous agent system with 643+ skills
Persistent Memory
Auto-generated memories and skills — it learns your projects and never forgets how it solved a problem. Cross-session conversation search.
Lives Where You Do
Telegram, Discord, Slack, WhatsApp, Signal, Email, CLI — start on one platform, pick up on another.
Subagent Parallelization
Spawn isolated subagents with their own conversations and terminals. Python RPC scripts for zero-context-cost pipelines.
Scheduled Automations
Natural language cron scheduling for reports, backups, and briefings — running unattended through the gateway.
Skills Learning Loop
643+ skills across 4 registries. The agent auto-creates skills after complex tasks and self-improves during use.
Real Sandboxing
Five backends — local, Docker, SSH, Singularity, Modal — with container hardening and namespace isolation.
Quick Install
Get started with Hermes Agent in three steps
Install
curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash Works on Linux, macOS, and WSL2. The installer handles everything — Python, Node.js, dependencies.
Configure
hermes setup Run the setup wizard to choose your model, configure platforms, and enable tools.
Start
hermes Start chatting! Supports CLI, Telegram, Discord, and more.
Requirements
- Linux / macOS / WSL2 (native Windows is not supported)
- Git
- Internet connection
- Recommended: at least one LLM API Key (OpenRouter, OpenAI, Anthropic, etc.)