HermitClaw
Micro autonomous AI agent that lives in a single folder. Folder-native workflow system with zero installation, portable configuration, and self-contained AI automation.
Micro Autonomous AI Agent - Lives in a Single Folder
HermitClaw is a micro autonomous AI agent designed around the concept of folder-native workflows. Everything lives in a single folder - configuration, skills, state, and logs. No installation required, just copy the folder and run. GitHub: https://github.com/brendanhogan/hermitclaw Developer: brendanhogan (Community Open Source) License: MIT Key Innovation: HermitClaw treats each AI workflow as a self-contained folder that can be easily shared, versioned, and deployed anywhere.
Core Philosophy: "One folder, one agent, zero hassle"
Why HermitClaw?
Folder-Native: Everything lives in one folder - copy, move, share, done.
Zero Installation: No system-wide changes, no package managers, just run.
Git-Friendly: Perfect structure for version control and collaboration.
Portable: Take your AI workflows anywhere on a USB drive.
Key Features
$#1.
Single Folder Architecture
my-agent/
├── config.yaml # Configuration
├── skills/ # AI skills and tools
├── workflows/ # Workflow definitions
├── state/ # Persistent state
├── logs/ # Execution logs
└── .env # Environment variables$#2.
Zero Installation
- No system packages required
- Works from any directory
- No admin privileges needed
- Instant deployment
$#3.
Workflow Definitions
- YAML-based workflow syntax
- Visual workflow editor (optional)
- Pre-built workflow templates
- Custom workflow creation
$#4.
Runtime Flexibility
- Python runtime support
- Node.js runtime support
- Choose per workflow
- Isolated dependencies
$#5.
Portability
- USB drive deployment
- Cloud sync compatible
- Easy backup and restore
- Cross-platform folders
Installation
Quick Start
# Clone or download
git clone https://github.com/brendanhogan/hermitclaw.git my-agent
cd my-agent
# Install dependencies (optional - can use system Python/Node)
pip install -r requirements.txt
# or
npm install
# Configure
cp .env.example .env
# Edit .env with your API keys
# Run
python main.py
# or
node index.jsCreate New Agent
# Copy template folder
cp -r hermitclaw-template my-new-agent
cd my-new-agent
# Customize config.yaml
# Add your workflows
# Run your agentConfiguration
config.yaml
name: my-hermit-agent
runtime: python # or nodejs
ai:
provider: openai
model: gpt-4o-mini
# api_key: from .env
workflows:
- name: daily-summary
schedule: "0 8 * * *"
enabled: true
storage:
state: ./state
logs: ./logs.env
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...Pricing
Free: HermitClaw is completely free and open source under MIT license. You only pay for the AI models you use through their respective providers.
System Requirements
| Component | Minimum | Recommended |
|---|---|---|
| OS | Any (Windows, macOS, Linux) | Any modern OS |
| Runtime | Python 3.8+ or Node.js 18+ | Python 3.11+ or Node.js 20+ |
| Memory | 256MB RAM | 512MB+ RAM |
| Storage | 50MB | 200MB |
| Network | Optional | Stable internet |
Use Cases
Personal Automation Scripts
Create personal AI workflows for daily tasks, reminders, and information gathering.
Portable AI Workflows
Carry your AI workflows on a USB drive and run them on any computer.
Team Workflow Sharing
Share workflow folders with team members via Git or file sharing.
Quick Prototyping
Rapidly prototype new AI workflows without complex setup.
Git-Based Versioning
Track workflow changes in Git, collaborate with team members.
Offline AI Tasks
Run AI workflows that process local files without internet.
Example Workflows
Daily News Summary
name: daily-news
trigger: schedule
schedule: "0 7 * * *"
steps:
- fetch: news_api
- summarize: ai
- output: emailFile Organizer
name: organize-downloads
trigger: folder_watch
folder: ~/Downloads
steps:
- scan: new_files
- classify: ai
- move: categorized_foldersCommunity and Support
- GitHub: https://github.com/brendanhogan/hermitclaw
- Documentation: https://github.com/brendanhogan/hermitclaw#readme
- Issues: GitHub Issues tab
- Examples: See examples/ folder in repository
FAQ
Is HermitClaw free to use?
Yes, HermitClaw is free and open source (MIT license). You only pay for AI model API costs if using external models.
What are the system requirements for HermitClaw?
HermitClaw requires 256MB RAM of RAM minimum. Runtime: Portable (Python or Node.js). It runs on Windows, macOS, and Linux.
Can I self-host HermitClaw?
Yes. HermitClaw is open source (MIT) and can be self-hosted on your own hardware. Clone the repository from GitHub and follow the installation guide.
How does HermitClaw compare to OpenClaw?
HermitClaw offers a different approach compared to OpenClaw. While OpenClaw provides the largest ecosystem with 13,729+ skills and maximum flexibility, HermitClaw focuses on personal assistant. Choose HermitClaw if you prioritize its specific features; choose OpenClaw for the broadest compatibility and community support.
Is HermitClaw suitable for beginners?
HermitClaw requires some technical knowledge to set up (Portable (Python or Node.js)). If you are a beginner, consider starting with QClaw (one-click install) or MaxClaw (cloud-based, no setup) first, then graduate to HermitClaw as you gain experience.
License
MIT License - Free for personal and commercial use.
Tags
ai-agent, portable, folder-native, zero-install, micro-agent, self-contained, yaml, git-friendly