- Git - Version control (checked during installation)
- Make - Build automation (checked during installation)
- Bash - Shell interpreter, version 3.2+ (checked during installation)
- GitHub CLI (
gh) - Required for GitHub integration features- Install: https://cli.github.com/
- Authenticate after installation:
gh auth login - Used by:
/setup-viewboard,/open-issue,/open-pr, GitHub workflow automation
- Python 3.10+ - Required for permission automation module, otherwise you can have infinite
yesto prompt!- Use Python
venvoranacondato manage a good Python release!
- Use Python
- Anthropic Python Library - For custom AI integrations (optional)
- Install:
pip install anthropic - Note: Not required for core SDK functionality, but recommended if you plan to extend or customize AI-powered features
- Install:
After installing prerequisites, the installer will automatically verify git, make, and bash availability. GitHub CLI authentication can be verified with:
gh auth statusAgentize is an AI-powered SDK that helps you build your software projects using Claude Code powerfully. It is splitted into two main components:
- Claude Code Plugin: Refer to Tutorial 00: Initialize Your Project to set up the Agentize plugin for Claude Code.
- CLI Tool: A source-first CLI tool to help you manage your projects using Agentize. See the commands below to install.
curl -fsSL https://raw.githubusercontent.com/SyntheSys-Lab/agentize/main/scripts/install | bashThen add to your shell RC file (~/.bashrc, ~/.zshrc, etc.):
source $HOME/.agentize/setup.shSee docs/cli/install.md for installation options and troubleshooting.
Upgrade: Run lol upgrade to pull the latest changes.
If you encounter any issue during the usage. For example:
- It asks you for permission on a really simple operation.
- It fails to automatically continue on a session.
export HANDSOFF_DEBUG=1Then re-run the command. This will give you a detailed log in either
/path/to/your/project/.tmp/handsoff-debug.logor$HOME/.agentize/.tmp/handsoff-debug.logPaste your logs on issue for me (@were) to debug!
For further help, please visit our troubleshooting guide.
- Plan first, code later: Use AI to generate a detailed plan before writing any code.
- Plan is put on Github Issues for tracking.
- Build skills.
- Skills are modular reusable, formal, and lightweighted flow definitions.
- This is something like C-style declaration and implementation separation.
/commandsare declarations and interfaces for users to invoke skills./skillsare implementations of the skills.
- Bootstrapping via self-improvement: We use
.claude/as our canonical rules directory. We use these rules to develop these rules further.- Top-down design: Start with a high-level view of the development flow.
- Bottom-up implementation: Implement each aspect of the flow from bottom, and finally integrate them together.
See our detailed workflow diagrams:
- Ultra Planner Workflow - Multi-agent debate-based planning
- Issue to Implementation Workflow - Complete development cycle
Legend: Red boxes represent user interventions (providing requirements, approving/rejecting results, starting sessions). Blue boxes represent automated AI steps.
Learn Agentize in 15 minutes with our step-by-step tutorials (3-5 min each):
- Initialize Your Project - Set up Agentize in new or existing projects
- You already did this if you followed the Quick Start!
- Ultra Planner - Primary planning tutorial (recommended)
- Issue to Implementation - Complete development cycle with
/issue-to-impland/code-review - Advanced Usage - Scale up with parallel development workflows
agentize/
├── .claude-plugin/ # Plugin root (use with --plugin-dir)
│ ├── marketplace.json # Plugin manifest
│ ├── commands/ # Claude Code commands
│ ├── skills/ # Claude Code skills
│ ├── agents/ # Claude Code agents
│ └── hooks/ # Claude Code hooks
├── python/ # Python modules (agentize.*)
├── docs/ # Documentation
│ └── git-msg-tags.md # Commit message conventions
├── src/cli/ # Source-first CLI libraries
│ ├── wt.sh # Worktree CLI library
│ └── lol.sh # SDK CLI library
├── scripts/ # Shell scripts and wrapper entrypoints
├── templates/ # Templates for SDK generation
├── tests/ # Test cases
├── Makefile # Build targets for testing and setup
└── README.md # This readme file