A mandatory meta-skill that enforces the invocation of relevant specialized skills before an AI agent takes any action or answers any question.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install using-superpowers
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 using-superpowers 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 Using Superpowers: AI Skill Orchestration?
The Using Superpowers skill serves as the foundational protocol for AI coding agents, ensuring they never bypass specialized workflows. By mandating a skill check at the very beginning of every interaction, it prevents agents from making assumptions or taking undisciplined actions that lead to errors or technical debt. It is a critical component of the Openclaw Skills ecosystem, designed to maximize the utility of specialized tools by removing the tendency of agents to rationalize their way out of following proven processes.
This skill acts as the primary entry point for all high-quality agentic reasoning. Whether an agent is faced with a simple query, a complex bug fix, or a request for codebase exploration, Using Superpowers ensures that the appropriate domain-specific Openclaw Skills are loaded and followed. It establishes a hierarchy of execution, prioritizing process-oriented skills like brainstorming and debugging over implementation-specific tasks.
Using Superpowers: AI Skill Orchestration Use Cases
- Initializing a new conversation with an AI coding agent to establish operational boundaries.
- Evaluating incoming user tasks to identify which specialized Openclaw Skills are required.
- Preventing agents from skipping disciplined workflows during simple clarifying questions.
- Ensuring multi-step tasks follow a consistent lifecycle of brainstorming, design, and execution.
How Using Superpowers: AI Skill Orchestration Works
- The agent receives a user message and immediately evaluates if any specialized skill might apply, even if the probability is as low as 1%.
- The agent invokes the Skill tool to load the relevant Openclaw Skills into its active context.
- The agent announces to the user specifically which skill is being used and for what purpose.
- If the loaded skill contains a checklist, the agent creates a Todo list via the TodoWrite tool for each item.
- The agent follows the skill instructions exactly, adhering to the rigid or flexible nature defined within the skill.
- Only after the skill protocol is initiated does the agent provide a response or ask clarifying questions.
Using Superpowers: AI Skill Orchestration Setup
To utilize this protocol within an environment like Claude Code, ensure the skill file is present in your skills directory. You can then invoke it or rely on the agent to trigger it based on its system instructions.
# Example of manually checking for skills in a CLI environment
skill using-superpowers
Ensure that the agent's environment is configured to allow the Skill tool to read and load other Openclaw Skills dynamically during the session.
Using Superpowers: AI Skill Orchestration Data Schema & Taxonomy
The Using Superpowers skill manages the metadata of session execution and skill prioritization. It organizes task flow according to the following taxonomy:
| Attribute | Description |
|---|---|
| Skill Priority | Ranks process-based skills (e.g., debugging) higher than implementation-based skills. |
| Red Flags | A set of logic checks used to identify when an agent is attempting to bypass protocols. |
| Skill Type | Classifies loaded Openclaw Skills as either Rigid (must follow exactly) or Flexible (adaptable). |
| Checklist State | Tracks the generation of Todo items based on the invoked skill's requirements. |
Using Superpowers: AI Skill Orchestration Advanced Features
- Anti-Rationalization Logic: Specifically designed to counter common agent excuses for skipping workflows, such as "this is just a simple question."
- Automated Checklist Generation: Automatically converts skill requirements into actionable tasks for the agent's internal todo manager.
- Dynamic Protocol Selection: Determines the correct order of operations when multiple Openclaw Skills apply to a single user request.
- Discipline Enforcement: Categorizes skills into Rigid and Flexible types to ensure that core engineering disciplines like TDD or debugging protocols are never compromised.
SKILL.md
Loading
Related Openclaw Skills
Using Git Worktrees
A specialized skill for creating and managing isolated Git workspaces with automated safety checks and environment setup.
Subagent-Driven Development
A high-velocity execution framework that dispatches specialized subagents for independent tasks with automated two-stage reviews.
Requesting Code Review
A specialized skill for triggering automated subagent code reviews to ensure quality and requirement alignment during development.
Finishing a Development Branch
A structured workflow for finalizing development work, ensuring all tests pass before automating merges or pull request creation.
Verification Before Completion
A rigorous technical protocol that mandates fresh evidence from tests and builds before any work is claimed as finished.
Writing Plans
A systematic blueprinting tool for AI agents to generate granular, TDD-focused implementation plans before writing code.
Featured*
