Back to AI Tools & Agents
using-superpowers
AI assistantskill invocationworkflow managementClaude CodeAI productivityagent skillsbest practices
Install this skill
npx skills add obra/superpowersWorks across Claude Code, Cursor, Codex, Copilot & Antigravity
The using-superpowers skill establishes a mandatory protocol for agents to identify and execute relevant task-specific workflows. It mandates that any agent operation—even minor queries—must undergo a rigorous evaluation against available skill sets before execution begins. If a 1% probability exists that a skill applies to the current request, the agent is strictly required to invoke that skill immediately. This creates a standardized, predictable environment where agents prioritize established best practices over impulsive, ad-hoc actions. By forcing a pre-task validation step, the system ensures that complex debugging, architecture, or coding tasks are handled through proven methodologies rather than individual guesswork. It eliminates subjective rationalizations like 'this is too simple for a skill' or 'I will check after looking at the code,' enforcing consistent discipline across all agent-led interactions.
When to Use This Skill
- •Commencing any new code generation task
- •Starting a debugging session for a failing test
- •Refactoring existing modules or components
- •Brainstorming architectural changes for a feature
How to Invoke This Skill
Example prompts that trigger this skill in Claude Code, Cursor, or Antigravity:
- “How should I start this feature?
- “Can you help me fix this error?
- “What is the best way to approach this refactor?
- “I need to add a new API endpoint
- “Let's build a authentication module
Pro Tips
- 💡Always invoke this 'superpowers' skill at the very beginning of any new interaction or task to establish the critical framework for skill usage.
- 💡Treat the '1% chance' rule as a mandatory diagnostic check for every user prompt; it's better to invoke and find a skill unnecessary than to miss a critical resource.
- 💡Familiarize yourself with the skill invocation mechanism in your specific environment (e.g., `Skill` tool in Claude Code) to ensure seamless and immediate application.
What this skill does
- •Forces mandatory pre-action skill evaluation
- •Categorizes tasks by process vs. implementation
- •Maps user intent to existing domain-specific toolkits
- •Provides a flow-based decision tree for task execution
- •Strictly bans bypass rationalization tactics
When not to use it
- ✕When the task is clearly outside the scope of any existing skills
- ✕If the request is purely administrative and unrelated to project work
Example workflow
- User submits a task request
- Agent performs a 1% probability check for skill applicability
- Agent invokes the required skill tool
- Agent announces the purpose of the skill usage
- Agent converts the skill checklist into individual tasks
- Agent executes the task by strictly following the skill steps
Prerequisites
- –A configured agent environment
- –Access to the agentpedia skills repository
Pitfalls & limitations
- !Over-invoking skills for irrelevant tasks might cause noise
- !Forgetting to check the latest version of a skill due to cognitive bias
- !Failing to prioritize process skills before implementation skills
FAQ
Do I really need to check for skills even for simple questions?
Yes. The protocol treats all questions as tasks. Checking for skills prevents errors by ensuring you are following established project conventions.
What if I feel like the skill is overkill for my current task?
You must invoke it anyway. The 1% rule is non-negotiable and intended to prevent rationalization that leads to poor outcomes.
How do I handle situations where multiple skills seem to apply?
Follow the Skill Priority order: handle process skills (like debugging or brainstorming) before implementation skills (like coding or design).
Should I try to gather info before invoking the skill?
No. The skill instruction explicitly states that the check comes before any information gathering or file exploration.
How it compares
This differs from manual task execution by enforcing an objective, rule-based filter that removes personal judgment, ensuring every task follows the team's vetted procedural standards.
📄 Full skill instructions — original source: obra/superpowers
<EXTREMELY-IMPORTANT>
If you think there is even a 1% chance a skill might apply to what you are doing, you ABSOLUTELY MUST invoke the skill.
IF A SKILL APPLIES TO YOUR TASK, YOU DO NOT HAVE A CHOICE. YOU MUST USE IT.
This is not negotiable. This is not optional. You cannot rationalize your way out of this.
</EXTREMELY-IMPORTANT>
## How to Access Skills
**In Claude Code:** Use the
**In other environments:** Check your platform's documentation for how skills are loaded.
# Using Skills
## The Rule
**Invoke relevant or requested skills BEFORE any response or action.** Even a 1% chance a skill might apply means that you should invoke the skill to check. If an invoked skill turns out to be wrong for the situation, you don't need to use it.
## Red Flags
These thoughts mean STOP—you're rationalizing:
| Thought | Reality |
|---------|---------|
| "This is just a simple question" | Questions are tasks. Check for skills. |
| "I need more context first" | Skill check comes BEFORE clarifying questions. |
| "Let me explore the codebase first" | Skills tell you HOW to explore. Check first. |
| "I can check git/files quickly" | Files lack conversation context. Check for skills. |
| "Let me gather information first" | Skills tell you HOW to gather information. |
| "This doesn't need a formal skill" | If a skill exists, use it. |
| "I remember this skill" | Skills evolve. Read current version. |
| "This doesn't count as a task" | Action = task. Check for skills. |
| "The skill is overkill" | Simple things become complex. Use it. |
| "I'll just do this one thing first" | Check BEFORE doing anything. |
| "This feels productive" | Undisciplined action wastes time. Skills prevent this. |
| "I know what that means" | Knowing the concept ≠ using the skill. Invoke it. |
## Skill Priority
When multiple skills could apply, use this order:
1. **Process skills first** (brainstorming, debugging) - these determine HOW to approach the task
2. **Implementation skills second** (frontend-design, mcp-builder) - these guide execution
"Let's build X" → brainstorming first, then implementation skills.
"Fix this bug" → debugging first, then domain-specific skills.
## Skill Types
**Rigid** (TDD, debugging): Follow exactly. Don't adapt away discipline.
**Flexible** (patterns): Adapt principles to context.
The skill itself tells you which.
## User Instructions
Instructions say WHAT, not HOW. "Add X" or "Fix Y" doesn't mean skip workflows.
If you think there is even a 1% chance a skill might apply to what you are doing, you ABSOLUTELY MUST invoke the skill.
IF A SKILL APPLIES TO YOUR TASK, YOU DO NOT HAVE A CHOICE. YOU MUST USE IT.
This is not negotiable. This is not optional. You cannot rationalize your way out of this.
</EXTREMELY-IMPORTANT>
## How to Access Skills
**In Claude Code:** Use the
Skill tool. When you invoke a skill, its content is loaded and presented to you—follow it directly. Never use the Read tool on skill files.**In other environments:** Check your platform's documentation for how skills are loaded.
# Using Skills
## The Rule
**Invoke relevant or requested skills BEFORE any response or action.** Even a 1% chance a skill might apply means that you should invoke the skill to check. If an invoked skill turns out to be wrong for the situation, you don't need to use it.
digraph skill_flow {
"User message received" [shape=doublecircle];
"Might any skill apply?" [shape=diamond];
"Invoke Skill tool" [shape=box];
"Announce: 'Using [skill] to [purpose]'" [shape=box];
"Has checklist?" [shape=diamond];
"Create TodoWrite todo per item" [shape=box];
"Follow skill exactly" [shape=box];
"Respond (including clarifications)" [shape=doublecircle];
"User message received" -> "Might any skill apply?";
"Might any skill apply?" -> "Invoke Skill tool" [label="yes, even 1%"];
"Might any skill apply?" -> "Respond (including clarifications)" [label="definitely not"];
"Invoke Skill tool" -> "Announce: 'Using [skill] to [purpose]'";
"Announce: 'Using [skill] to [purpose]'" -> "Has checklist?";
"Has checklist?" -> "Create TodoWrite todo per item" [label="yes"];
"Has checklist?" -> "Follow skill exactly" [label="no"];
"Create TodoWrite todo per item" -> "Follow skill exactly";
}## Red Flags
These thoughts mean STOP—you're rationalizing:
| Thought | Reality |
|---------|---------|
| "This is just a simple question" | Questions are tasks. Check for skills. |
| "I need more context first" | Skill check comes BEFORE clarifying questions. |
| "Let me explore the codebase first" | Skills tell you HOW to explore. Check first. |
| "I can check git/files quickly" | Files lack conversation context. Check for skills. |
| "Let me gather information first" | Skills tell you HOW to gather information. |
| "This doesn't need a formal skill" | If a skill exists, use it. |
| "I remember this skill" | Skills evolve. Read current version. |
| "This doesn't count as a task" | Action = task. Check for skills. |
| "The skill is overkill" | Simple things become complex. Use it. |
| "I'll just do this one thing first" | Check BEFORE doing anything. |
| "This feels productive" | Undisciplined action wastes time. Skills prevent this. |
| "I know what that means" | Knowing the concept ≠ using the skill. Invoke it. |
## Skill Priority
When multiple skills could apply, use this order:
1. **Process skills first** (brainstorming, debugging) - these determine HOW to approach the task
2. **Implementation skills second** (frontend-design, mcp-builder) - these guide execution
"Let's build X" → brainstorming first, then implementation skills.
"Fix this bug" → debugging first, then domain-specific skills.
## Skill Types
**Rigid** (TDD, debugging): Follow exactly. Don't adapt away discipline.
**Flexible** (patterns): Adapt principles to context.
The skill itself tells you which.
## User Instructions
Instructions say WHAT, not HOW. "Add X" or "Fix Y" doesn't mean skip workflows.
How to Use This Skill Unit
Option A: Project-Specific (Recommended)
- Click "Download" above
- In your project, create the directory:
.agent/skills/using-superpowers/ - Save the file as
SKILL.md - The agent will automatically discover the skill based on its description.
Option B: Global Installation (All Agents)
Save the file to these locations to make it available across all projects:
- Claude Code:
~/.claude/skills/obra/superpowers/using-superpowers/SKILL.md - Cursor:
~/.cursor/skills/obra/superpowers/using-superpowers/SKILL.md - Antigravity:
~/.gemini/antigravity/skills/obra/superpowers/using-superpowers/SKILL.md
🚀 Install with CLI:npx skills add obra/superpowers
Recommended Rules
View more rules →Go Fundamentals & Best Practices
GoGolangBest Practices
You are an expert in Go (Golang) programming and best practices. Key Principles: - Follow idiomatic Go (Effective Go) - Keep it simple and readable -...
Unit Testing Best Practices
TestingUnit TestQA
You are an expert in Unit Testing principles and best practices. Key Principles: - Test small, isolated units of code (functions/classes) - Tests mus...
React Hooks Best Practices
ReactHooksBest Practices
You are an expert in React Hooks. Key Principles: - Follow Rules of Hooks strictly - Use custom hooks for reusable logic - Optimize dependency arrays...
Recommended Workflows
View more workflows →Fix Next.js Hydration Errors
Next.jsDebuggingHydration
--- description: Systematically debug and fix 'Text content does not match server-rendered HTML' errors --- 1. **Check for Invalid HTML Nesting**: ...
Nuke & Reinstall
npmTroubleshootingDependencies
--- description: The nuclear option for when dependencies are completely broken --- 1. **Remove node_modules**: - Delete the existing `node_module...
Debugging Infinite Re-renders
ReactDebuggingPerformance
--- description: Track down and fix infinite loops in useEffect and component rendering --- 1. **Check `useEffect` Dependencies**: - The most comm...
