A 4-phase agile framework for AI coding agents to manage the full software development lifecycle from analysis to implementation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install lb-bmad-skill
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install lb-bmad-skill using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is BMad Method?
BMad (Breakthrough Method of Agile AI Driven Development) is a comprehensive framework designed to optimize the performance of autonomous AI agents. It structures the development process into four distinct phases: Analysis, Planning, Solutioning, and Implementation. By integrating this framework with Openclaw Skills, developers can transform high-level project goals into technical documentation and executable code with high precision.
The skill relies on Claude Code to execute complex workflows, ensuring that every piece of generated code is backed by a solid architectural foundation. It utilizes a project-context.md file as the project's brain, allowing agents to maintain long-term context and adhere to specific technical standards across multiple sessions.
BMad Method Use Cases
- Designing full-stack architectures and Architectural Decision Records (ADRs) for new projects.
- Breaking down complex product epics into granular, implementation-ready user stories.
- Automating sprint planning and status tracking via structured YAML and Markdown outputs.
- Rapidly implementing UI tweaks and non-critical features using the Quick-Dev workflow.
- Maintaining consistent project documentation including PRDs and UX specifications.
How BMad Method Works
- Verification: The agent performs a pre-flight check to ensure BMad and the required project-context.md are present.
- Discovery: Analysis commands explore the codebase and problem space to define the project scope.
- Definition: The agent generates product requirements (PRDs) and technical specifications based on the discovery phase.
- Architecture: Solutioning commands create the technical blueprints and select appropriate AI models for implementation.
- Execution: The framework generates specific stories which are then implemented by coding agents or sub-agents with full context traceability.
BMad Method Setup
To initialize the BMad framework within your project directory, ensure you have the coding-agent skill and Claude Code installed, then run the interactive installer:
cd ~/project && npx bmad-method install
Follow the prompts to select Claude Code as your primary AI tool and install the necessary modules. After installation, verify the setup by checking for the internal configuration directory:
ls -la ~/project/_bmad/
BMad Method Data Schema & Taxonomy
The skill organizes all generated artifacts within a structured directory to maintain traceability and context for Openclaw Skills agents.
| Directory/File | Description |
|---|---|
project-context.md |
The persistent context file acting as the project's brain. |
_bmad-output/ |
The primary storage for all generated documentation and reports. |
_bmad-output/PRD.md |
High-level product requirement documentation. |
_bmad-output/architecture.md |
Detailed technical architecture and decision logs. |
_bmad-output/epics/ |
Contains sub-directories for each epic and its associated user stories. |
_bmad-output/sprint-status.yaml |
A machine-readable file tracking the current implementation progress. |
BMad Method Advanced Features
- Sequential Command Chaining: Robust workflow patterns that verify the output of one phase before starting the next.
- Context Injection for Sub-Agents: Specialized prompts that provide minions with architecture, context, and story files simultaneously.
- Permission Bypass Mode: Support for non-interactive execution using flags like --dangerously-skip-permissions for automated pipelines.
- Crash Recovery Protocols: Standardized steps to clean zombie processes and resume workflows from the last valid generated file.
- Model Specialization: Strategic routing of tasks between Sonnet, Haiku, and Opus based on task complexity and token efficiency.
SKILL.md
Loading
Related Openclaw Skills
Better Auth Documentation Integration
A comprehensive technical reference skill for implementing modern, secure authentication in TypeScript projects using AI agents.
Xiangji Translation Service
Xiangji Translation Service provides professional-grade text and image translation capabilities by integrating multiple global AI engines.
Multi-LLM Intelligent Switching
An intelligent model routing tool that switches between high-performance cloud LLMs and local Ollama instances based on task intent and user commands.
AutoFillIn
An intelligent browser automation skill for filling complex web forms and uploading files with persistent login support.
Motion.dev Animation Library
A high-performance animation skill for implementing Motion.dev (formerly Framer Motion) across React, Vue, and vanilla JavaScript projects.
Pocket-TTS
Pocket-TTS is a lightweight, CPU-optimized text-to-speech engine that enables real-time streaming and voice cloning without requiring a GPU.
METADATA
Featured*
