Skills Directory

Performs deep, structured code reviews covering correctness, security, performance, maintainability, and testing gaps. Outputs actionable patches and prioritized findings.

Audits code and configuration for vulnerabilities: injection, broken auth, secrets, dependency risks, and insecure defaults. Reports severity-ranked findings with fixes.

Finds and fixes real performance problems: N+1 queries, unnecessary renders, slow hot paths, and memory growth. Measures before and after — no cargo-cult optimization.

Writes comprehensive, maintainable unit, integration, and property-based tests. Focuses on high-value coverage and realistic edge cases.

Systematically diagnoses bugs and production issues. Reproduces the problem, narrows the root cause, and proposes minimal fixes with explanations.

Safely performs coordinated refactors across many files while preserving behavior. Produces a step-by-step plan + the exact edits.

Designs clean, evolvable APIs (REST, GraphQL, or RPC). Produces OpenAPI specs, client code, error models, and migration guides.

Upgrades dependencies safely: reads changelogs, identifies breaking changes, migrates code, and verifies with the test suite — one risk group at a time.

Generates excellent, accurate technical documentation from source code, types, and usage examples. Produces README sections, API docs, and guides.

Creates conventional commits, clear PR descriptions, and release notes. Handles branching strategy and changelog generation.

Explains unfamiliar code at the right altitude: architecture first, then flow, then details. Perfect for onboarding, code archaeology, and "what does this actually do?"

Prepares optimal context for large codebases or agent sessions. Builds smart indexes, summaries, and retrieval strategies so agents stay within token limits while remaining effective.

Improves prompts, skills, and AGENTS.md files using a measure-first loop. Diagnoses why an instruction fails and rewrites it to be reliable.

Anthropic's official skill for creating, editing, and analyzing Word documents — with tracked changes, comments, and formatting preservation.

Anthropic's official PDF skill: extract text and tables, fill forms, merge/split, and generate new PDFs — programmatically, not by guessing.

Anthropic's official PowerPoint skill: build and edit presentations with layouts, templates, charts, and speaker notes.

Anthropic's official Excel skill: read and write spreadsheets with formulas, formatting, multiple sheets, and data analysis.

Anthropic's official guide skill for building high-quality MCP servers that wrap external APIs cleanly.

Anthropic's official interactive skill for authoring new skills — scaffolds SKILL.md, structure, and trigger descriptions the right way.

Anthropic's official skill for testing web apps with Playwright — drive the UI, verify behavior, and capture what actually rendered.

Anthropic's official reference skill for the Claude API & SDKs: model IDs, pricing, streaming, tool use, MCP, caching, and migration.

Anthropic's official generative-art skill: create visual art with p5.js using randomness, particle systems, and generative techniques.

Anthropic's official skill for creating polished visual designs as PNG and PDF, grounded in real design principles.

Anthropic's official skill for making animated GIFs tuned to Slack's size and dimension constraints.

Anthropic's official skill for applying a consistent brand — colors, typography, and voice — across generated artifacts.

Anthropic's official skill for escaping generic AI aesthetics in React + Tailwind — distinctive, intentional UI instead of cookie-cutter.

Anthropic's official skill for building rich HTML artifacts with React, Tailwind, and shadcn/ui.

Anthropic's official skill for styling any artifact with a theme — 10 preset color/font themes, or generate a custom one on the fly.

Anthropic's official skill for writing clear internal communications: status reports, newsletters, announcements, and FAQs.

Anthropic's official skill for co-authoring substantial docs — specs, proposals, decision docs — through a structured, reader-tested workflow.

Enforces a strict RED → GREEN → REFACTOR loop with an anti-patterns reference. The flagship skill from the Superpowers library.

A disciplined 4-phase root-cause process so the agent stops guess-and-check debugging. From the Superpowers library.

Forces the agent to prove a change works before claiming it's done. Kills the #1 agent failure mode. From the Superpowers library.

A pre-review validation checklist the agent runs before asking for review, so reviewers get clean diffs. From the Superpowers library.

A process for responding to review feedback well — triage, address, and confirm — instead of blindly applying every comment. From Superpowers.

Refines vague ideas into clear designs through Socratic questioning before any code is written. From the Superpowers library.

Produces detailed, reviewable implementation plans with concrete tasks before execution. From the Superpowers library.

Executes an approved plan in batches with human checkpoints, so long tasks stay on the rails. From the Superpowers library.

Sets up and uses git worktrees so the agent can run parallel branches of work without clobbering your checkout. From Superpowers.

Guides the end-of-branch decision: merge, PR, or clean up — with the right verification first. From the Superpowers library.

Rapid iteration using subagents with a two-stage review loop, keeping the main context clean. From the Superpowers library.

Coordinates multiple concurrent subagents safely — splitting work and merging results. From the Superpowers library.

General-purpose browser automation as a skill: navigate, interact, and assert with Playwright. Popular community skill.

Build, launch, navigate, and test iOS apps in the simulator through automation. Community skill for mobile work.

Authorized web fuzzing guidance with ffuf — content discovery and parameter fuzzing for security testing. Community skill.

Professional security-audit skills from Trail of Bits: static analysis with CodeQL/Semgrep, variant analysis, and code auditing.

Build real, data-bound visualizations in D3.js — proper scales, axes, and transitions, not chart-library defaults. Community skill.

Generate favicons, app icons, and social/OG images at the right sizes and formats. Community skill for shipping web projects.

Create animation-rich HTML presentations (reveal.js-style) instead of static decks. Community skill.

Official Expo team skills for building React Native apps with Expo — routing, native modules, and EAS workflows.