Hermes Agent - GitHub README

Source: https://github.com/NousResearch/hermes-agent

Overview

Hermes Agent is an autonomous AI system developed by Nous Research that operates across multiple platforms. It features a built-in learning mechanism, persistent memory systems, and the ability to function on various infrastructure setups -- from modest $5 VPS deployments to GPU clusters and serverless environments.

Core Features

**Self-Improving Architecture** The agent creates and refines skills autonomously, maintains conversational history with full-text search capabilities, and develops user profiles that persist across sessions.

**Multi-Platform Access** "Telegram, Discord, Slack, WhatsApp, Signal, and CLI -- all from a single gateway process." The system supports voice transcription and maintains conversation continuity across platforms.

**Terminal Interface** A comprehensive TUI offering multiline editing, command autocomplete, conversation history, interrupt functionality, and streaming tool output.

**Flexible Model Support** Compatible with numerous providers including Nous Portal, OpenRouter (200+ models), NovitaAI, NVIDIA NIM, OpenAI, and custom endpoints. Model switching occurs without code modifications.

**Scheduled Automation** Built-in cron scheduler enabling natural-language task automation for recurring operations like daily reports or weekly audits.

**Distributed Execution** Capability to spawn isolated subagents for parallel workstreams and collapse multi-step Python scripts into zero-context-cost operations.

Installation

**Linux/macOS/WSL2/Termux:**

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

**Windows (PowerShell):**

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

The Windows installer provides bundled Git Bash (MinGit), Python 3.11, Node.js, ripgrep, and ffmpeg without requiring administrator privileges.

Quick Start Commands

• hermes -- Launch interactive CLI
• hermes model -- Select LLM provider
• hermes tools -- Configure enabled tools
• hermes gateway -- Start messaging platform bridge
• hermes setup -- Full configuration wizard
• hermes update -- Update to latest version

Messaging & CLI Reference

Technical Stack

**Languages:** Python (83.6%), TypeScript (12.6%), JavaScript (1.0%)

**Key Components:**

• Agent loop with tool-calling architecture
• FTS5 session search with LLM summarization
• Memory system compatible with Honcho user modeling
• Skills standard aligned with agentskills.io
• MCP (Model Context Protocol) integration support
• Terminal backends: local, Docker, SSH, Singularity, Modal, Daytona

Migration & Compatibility

OpenClaw users can automatically import configurations via hermes claw migrate, transferring SOUL.md personas, memories, skills, command allowlists, and API keys.

Development

**Quick contributor setup:**

git clone https://github.com/NousResearch/hermes-agent.git
cd hermes-agent
./setup-hermes.sh

Community & Resources

• Discord server for community discussion
• Skills Hub at agentskills.io for procedural memory sharing
• GitHub issues for bug reports
• Comprehensive documentation at hermes-agent.nousresearch.com/docs

Licensing

MIT License -- Open source and freely distributable.