Use for any code change — bug fix, small feature, refactor, or improvement. Ensures TDD, clarifying questions, codebase investigation, DRY, proper branching/worktrees, UI skills, and doc updates. PDD invokes this automatically for large features.
Use when fixing a bug — enforces reproduce-first-then-fix discipline with a strict step-by-step protocol. Prevents wrong-approach loops by gating each step on success criteria.
Use after completing a feature or bug fix to log the change in CHANGELOG.md using plain, business-user-friendly language
Reconcile shipped code with the plan in both directions — additive drift (unplanned work that landed) and subtractive drift (dead files, stale TODOs, completed deferred items). Run as part of PDD's post-completion audit, before /simplify.
Generate 5 meaningfully distinct UI/UX variants of the same screen in parallel (one subagent each), reachable via ?style=1...5, so the user can compare and pick. Calling again on a chosen style refines within that direction.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Reusable Claude Code skills for disciplined software development.
| Skill | Description |
|---|---|
| best-practices | The baseline for every code change. Branching (never main), investigate before building, ask before assuming, TDD, DRY, simplify after implementation, update-docs when done, lint before committing. |
| test-driven-development | Write failing tests first against real infrastructure. No mocks. Red-green-refactor. |
| plan-driven-development | Versioned plan files in plans/ for multi-milestone features. Builds on best-practices. Skill audit before planning, scenario map for complex changes, plan-vs-reality audit before PR, multi-agent coordination across context windows. |
| user-research | Product-manager-grade discovery before building a user-facing feature — problem statement, jobs-to-be-done, core flow, and a decision matrix. Output is a short brief the requester approves, with open decisions surfaced as explicit questions. Invoked by plan-driven-development and give-me-five. |
| bug-fix | Strict 10-step protocol for fixing bugs. Reproduce first, then fix. Each step gates on success criteria. Supports Chrome browser debugging for UI bugs. |
| change-log | Log each feature or fix in CHANGELOG.md using plain, business-user-friendly language. Powers the product updates blog. |
| simple-design-principles | Rules for user-facing copy and UI components. Plain language, no jargon, consistent component usage. |
| give-me-five | Generate 5 meaningfully distinct UI/UX variants of the same screen in parallel — one subagent per variant — each reachable via ?style=1...5. Iterate on a chosen style to refine within that direction. |
| self-documenting-apis | Ensure FastAPI endpoints have docstrings, typed response/request models, and proper status codes so auto-generated docs are the single source of truth. |
| update-docs | Sync all documentation after completing a feature. Finds stale content across CLAUDE.md, README, public skills/plugins, and downstream project docs. |
| show-me | Prove a change works after implementing it — derive a minimal test plan from the diff, run the app, drive it in the browser with an animated cursor and styled step captions, and deliver a short MP4 of just the page viewport (or terminal transcript for non-UI changes) as proof. Inspired by Devin's testing-and-recordings workflow. |
| drift-audit | Run before the PDD post-completion audit. Reconciles shipped code vs the plan in both directions — additive drift (unplanned things that landed) and subtractive drift (orphan files, completed Post-MVP items, stale TODOs, dead redirect stubs). |
| Quick-question mode. Ultra-short answers and drafts. Drafted texts/messages and emails fit under 480 characters — no greetings, sign-offs, or preamble. |
In Claude Code, run:
/plugin marketplace add jackyliang/powerups
/plugin install powerups@powerups
MIT
npx claudepluginhub jackyliang/powerups --plugin powerupsCore developer skills for feature planning, code review, testing, commits, and daily development workflows.
Test-driven development.
Helder's personal SDLC toolbelt for AI coding agents — from spec to ship. Bundles the tracer-bullet workflow alongside TDD, code review, audits, and shipping skills.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Production-grade engineering skills for AI coding agents — covering the full software development lifecycle from spec to ship.
Ryan's riff on obra/superpowers: strong guidance for Claude Code as a software development assistant