.png)
4 weeks ago
1
Give Claude Code superpowers with a comprehensive skills library of proven techniques, patterns, and tools.
The superpowers plugin is a minimal shim that:
- Clones/updates the superpowers-skills repository to ~/.config/superpowers/skills/
- Registers hooks that load skills from the local repository
- Offers users the option to fork the skills repo for contributions
All skills, scripts, and documentation live in the separate superpowers-skills repository. Edit skills locally, commit changes, and contribute back via pull requests.
Skills Repository: https://github.com/obra/superpowers-skills
- Testing Skills - TDD, async testing, anti-patterns
- Debugging Skills - Systematic debugging, root cause tracing, verification
- Collaboration Skills - Brainstorming, planning, code review, parallel agents
- Meta Skills - Creating, testing, and contributing skills
Plus:
- Slash Commands - /brainstorm, /write-plan, /execute-plan
- Skills Search - Grep-powered discovery of relevant skills
- Gap Tracking - Failed searches logged for skill creation
Read the introduction: Superpowers for Claude Code
The plugin automatically handles skills repository setup on first run.
The plugin fetches and fast-forwards your local skills repository on each session start. If your local branch has diverged, Claude notifies you to use the pulling-updates-from-skills-repository skill.
If you forked the skills repository during setup, you can contribute improvements:
- Edit skills in ~/.config/superpowers/skills/
- Commit your changes
- Push to your fork
- Open a PR to obra/superpowers-skills
Find skills before starting any task:
Brainstorm a design:
Create an implementation plan:
Execute the plan:
Testing (skills/testing/)
- test-driven-development - RED-GREEN-REFACTOR cycle
- condition-based-waiting - Async test patterns
- testing-anti-patterns - Common pitfalls to avoid
Debugging (skills/debugging/)
- systematic-debugging - 4-phase root cause process
- root-cause-tracing - Find the real problem
- verification-before-completion - Ensure it's actually fixed
- defense-in-depth - Multiple validation layers
Collaboration (skills/collaboration/)
- brainstorming - Socratic design refinement
- writing-plans - Detailed implementation plans
- executing-plans - Batch execution with checkpoints
- dispatching-parallel-agents - Concurrent subagent workflows
- remembering-conversations - Search past work
- using-git-worktrees - Parallel development branches
- requesting-code-review - Pre-review checklist
- receiving-code-review - Responding to feedback
Meta (skills/meta/)
- writing-skills - TDD for documentation, create new skills
- sharing-skills - Contribute skills back via branch and PR
- testing-skills-with-subagents - Validate skill quality
- pulling-updates-from-skills-repository - Sync with upstream
- gardening-skills-wiki - Maintain and improve skills
- brainstorm.md - Interactive design refinement using Socratic method
- write-plan.md - Create detailed implementation plans
- execute-plan.md - Execute plans in batches with review checkpoints
- find-skills - Unified skill discovery with descriptions
- skill-run - Generic runner for any skill script
- search-conversations - Semantic search of past Claude sessions (in remembering-conversations skill)
Using tools:
- SessionStart Hook - Clone/update skills repo, inject skills context
- Skills Discovery - find-skills shows all available skills with descriptions
- Mandatory Workflow - Skills become required when they exist for your task
- Gap Tracking - Failed searches logged for skill development
- Test-Driven Development - Write tests first, always
- Systematic over ad-hoc - Process over guessing
- Complexity reduction - Simplicity as primary goal
- Evidence over claims - Verify before declaring success
- Domain over implementation - Work at problem level, not solution level
MIT License - see LICENSE file for details
