A disciplined engineering framework for Openclaw Skills that utilizes sub-agents and a rigorous 5-step lifecycle to deliver high-quality code.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install elcano-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 elcano-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 Superpowers: Structured Multi-Agent Development?
Superpowers is a specialized development protocol adapted for Openclaw Skills that transforms the AI agent into a structured engineering team. It moves away from impulsive coding by enforcing a strict hierarchy of operations: Brainstorm, Plan, Execute, Review, and Ship. By leveraging sub-agent spawning, it ensures that every task is performed in a clean environment, preventing the context contamination that often plagues long-running AI development sessions.
This skill is designed for complex engineering tasks such as large-scale refactors or multi-file feature additions. It integrates principles like Test-Driven Development (TDD), YAGNI (You Ain't Gonna Need It), and DRY (Don't Repeat Yourself) into the core logic of the Openclaw Skills workflow. By requiring design approval before any code is written, Superpowers reduces technical debt and ensures that the final output aligns perfectly with project requirements.
Superpowers: Structured Multi-Agent Development Use Cases
- Implementing complex new features that span multiple components or files.
- Executing major refactors where architectural integrity is a high priority.
- Maintaining high test coverage through mandatory red-green-refactor cycles.
- Managing high-stakes production environments where manual verification and screenshots are required.
- Standardizing development workflows across different contributors using Openclaw Skills.
How Superpowers: Structured Multi-Agent Development Works
- Brainstorming: The agent explores the current codebase, proposes 2-3 architectural approaches, and saves an approved design to the docs directory.
- Planning: A granular implementation plan is generated, breaking down the feature into 2-5 minute tasks with exact file paths and test cases.
- Execution: The system uses session spawning to create independent sub-agents for each task, ensuring isolation and fresh context.
- Dual Review: Every task is reviewed for both specification compliance and general code quality before proceeding.
- Shipping: Final verification includes build checks, UI validation via screenshots, and automated deployment notifications.
Superpowers: Structured Multi-Agent Development Setup
To activate Superpowers within your environment, ensure your project directory is initialized for Openclaw Skills and create the necessary plan storage.
mkdir -p docs/plans
Configure your agent to reference the Superpowers protocol whenever a new feature or refactor is requested. It is recommended to use high-reasoning models like Claude 3.5 Sonnet for the sub-agent implementation tasks.
Superpowers: Structured Multi-Agent Development Data Schema & Taxonomy
The skill organizes its engineering intelligence through a structured documentation hierarchy and session metadata.
| Location | Type | Description |
|---|---|---|
docs/plans/YYYY-MM-DD-<name>-design.md |
Markdown | Architectural decisions, data flows, and trade-offs. |
docs/plans/YYYY-MM-DD-<name>-plan.md |
Markdown | Granular task list, full code snippets, and test expectations. |
sessions_spawn |
Metadata | Contextual parameters for sub-agent isolation. |
git commit |
Version Control | Standardized feat/refactor/test commit messages for every task. |
Superpowers: Structured Multi-Agent Development Advanced Features
- Sub-Agent Task Isolation: Uses session spawning to prevent context drift during long-form development within Openclaw Skills.
- Mandatory TDD Gates: Sub-agents are hard-coded to write failing tests before any implementation logic is allowed.
- Spec Compliance Auditing: An automated review phase that reverts any code changes not explicitly defined in the implementation plan.
- Automated UI Verification: Built-in hooks for restarting services like PM2 and capturing screenshots for visual verification.
- Multi-Agent Orchestration: Orchestrates the flow between a high-level architect agent and atomic implementer agents.
SKILL.md
Loading
Related Openclaw Skills
Cline AI Coding CLI
Cline is a command-line autonomous agent that allows developers to plan, build, and refactor code through natural language instructions.
ResearchMonitor
ResearchMonitor is an automated agent skill that tracks new academic papers, journals, and conference announcements tailored to your specific research interests.
Fitness Coach
A professional-grade, dual-persona coaching system that delivers science-backed fitness and nutrition guidance through evidence-based research.
Clawnet
Clawnet is a stealth Chromium browser skill for AI agents that automates bot bypass, residential proxy management, and CAPTCHA solving.
Payroll Oracle
A decentralized HR and payroll infrastructure that audits GitHub or Linear work to trigger automated USDC payments via the x402 protocol.
ClawHub Intro Skill
A foundational demonstration skill designed to showcase the standardized structure and publication workflow for the Openclaw registry.
Featured*
