A composable skills framework that provides coding agents with structured development workflows including TDD, systematic debugging, and subagent-driven execution.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install obra-superpowers-agentic-workflow
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 obra-superpowers-agentic-workflow 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 obra/superpowers Agentic Skills Framework?
The Superpowers Agentic Skills Framework is designed to transform standard AI coding assistants into highly disciplined software engineers. By integrating these Openclaw Skills into your development environment, you ensure that your AI agent follows industry best practices without constant manual prompting. The framework enforces a design-first approach where Socratic brainstorming and technical planning precede any code generation, resulting in more robust and maintainable software architecture.
Whether you are using Claude Code, Cursor, Codex, or Gemini CLI, this framework bridges the gap between raw LLM output and professional-grade development. It manages the entire lifecycle of a feature, from initial design to the final merge, all while maintaining strict adherence to Red-Green-Refactor methodologies and isolated development environments via git worktrees.
obra/superpowers Agentic Skills Framework Use Cases
- Implementing complex new features with mandatory design-first planning requirements.
- Automating Test-Driven Development (TDD) cycles to ensure 100% test coverage for new logic.
- Managing parallel feature development using isolated git worktrees to keep the main branch clean.
- Scaling development throughput via subagent-driven execution for multi-task workflows.
- Troubleshooting regression bugs through a structured 4-phase systematic debugging process.
How obra/superpowers Agentic Skills Framework Works
- The agent automatically triggers the brainstorming skill to extract technical specifications and secure design approvals before writing code.
- A dedicated git worktree is created to isolate the feature development branch from the main workspace.
- The agent generates a granular implementation plan where every task includes exact file paths, code snippets, and verification steps.
- Fresh subagents are dispatched to execute individual tasks, each undergoing a two-stage review for spec compliance and quality.
- A strict TDD cycle is enforced for every task, requiring a failing test (RED) before minimal code is written to pass (GREEN).
- The framework conducts a final code review against the original plan and quality standards.
- Upon successful completion, the agent presents merge options and automatically cleans up the temporary worktree.
obra/superpowers Agentic Skills Framework Setup
To install these Openclaw Skills on Claude Code, use the official plugin command:
/plugin install superpowers@claude-plugins-official
For Cursor users, simply request the plugin in the agent chat:
/add-plugin superpowers
For Gemini CLI, run the following extension installation command:
gemini extensions install https://github.com/obra/superpowers
Verify the installation by starting a new session and asking your agent to help you plan a feature; the brainstorming skill should activate automatically.
obra/superpowers Agentic Skills Framework Data Schema & Taxonomy
The framework organizes its logic into specific skill directories and maintains metadata for workflow triggers. Data is structured as follows:
| Component | Description |
|---|---|
skills/<name>/SKILL.md |
Contains the name, description, and triggers for each agentic capability. |
| Git Worktrees | Isolated directories created for parallel development branches. |
| Implementation Plans | Markdown-formatted task lists containing file paths, tests, and expected outcomes. |
| Review Logs | Structured feedback categorized by severity (critical, major, minor). |
obra/superpowers Agentic Skills Framework Advanced Features
- Multi-agent orchestration for concurrent feature development and parallel task execution.
- Anti-pattern detection that automatically deletes production code written before a corresponding test exists.
- Custom skill creation engine that allows developers to extend the framework with new Openclaw Skills.
- Automated git worktree lifecycle management to prevent workspace pollution during multi-tasking.
- Systematic 4-phase debugging workflow including reproduction, root cause analysis, fix implementation, and verification.
SKILL.md
Loading
Related Openclaw Skills
NVIDIA NemoClaw
A secure orchestration plugin for running always-on AI assistants in policy-enforced sandboxes using NVIDIA OpenShell.
Agent Architecture Evaluator
A comprehensive diagnostic tool for reviewing, testing, and optimizing the structural integrity of complex agentic architectures and multi-agent systems.
pidrive
A CLI-based storage solution that mounts AWS S3 as a local filesystem via WebDAV, enabling AI agents to manage files using standard Unix commands.
SkillBoss Multi-Model AI Hub
Access 50+ AI models for chat, media generation, and developer tools via a single API key through Openclaw Skills.
ProductClank Campaigns
Turn your community into a growth engine with AI-powered social media amplification and authentic conversation discovery.
Second Level Thinking Investment Analyst
A sophisticated investment analysis skill that applies Howard Marks' framework to challenge market consensus and identify mispriced assets.
Featured*
