AI coding & development skills for agents

Helps design stable, well-documented APIs and interfaces with consistent contracts, error handling, validation at boundaries, and backward-c

Refactors working code to be clearer and easier to maintain while preserving exact behavior and following project conventions.

Provides backend architecture patterns, API design, database optimization, and server-side best practices for Node.js, Express, and Next.js

Designs and reviews REST API endpoints with best practices for URLs, status codes, pagination, and more.

Performs multi-axis code review across correctness, readability, architecture, security, and performance before any change is merged.

Automates a headless Chromium browser via the Playwright MCP server to navigate, click, type, screenshot, and verify web pages including loc

Uses Chrome DevTools MCP to inspect live browser behavior—DOM, console, network, performance, and screenshots—to debug and verify frontend c

Systematically identifies and fixes the root cause of errors in code.

Manages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to ano

Ensures every significant decision is rigorously challenged before implementation.

Guides completing a development branch by verifying tests, then merging locally, pushing a PR, or keeping the branch, with safe worktree cle

Manages code changes through structured Git workflows, atomic commits, and versioning practices.

Delivers changes incrementally in thin, verifiable slices. Use when implementing any feature or change that touches more than one file, or w

Optimizes application performance across frontend, backend, queries, and databases. Use when performance requirements exist, when you suspec

Helps evaluate and respond to code review feedback with technical verification and reasoned pushback instead of blind implementation or perf

Verifies every framework-specific implementation decision against official documentation before writing code, citing sources and avoiding ou

Guides writing and approving a structured specification with objectives, boundaries, and success criteria before any code is implemented.

Enforces test-driven development by writing a failing test first, then minimal implementation code, then refactoring, for every feature or b

Guides development using the test-driven red-green-refactor loop, writing failing tests first to prove code works and reproduce bugs before

Sets up an isolated git worktree workspace before starting feature work or implementing plans, including detection, creation, dependency set

Turns a spec or requirements into a detailed, test-driven implementation plan with bite-sized tasks before any code is written.

Dispatches a code reviewer subagent with crafted context to verify completed work meets requirements before merging or proceeding.

Enforces running actual verification commands (tests, builds, linters) and checking their output before making any claims that work is compl

Automates upgrading Remotion libraries, dependencies, and related packages in a project.

Establishes a project's quality bar as a written contract and stops agents quietly lowering it. Interviews the user on which dimensions matt

Guidance on using Bun as a JavaScript runtime, package manager, bundler, and test runner, including when to choose it over Node and how to m

Creating or modifying adapters for AI agents across server, UI, and CLI components.

A developer guide for adding a new issue-thread interaction card kind end-to-end in the Paperclip codebase, covering shared contracts, serve

A structured debugging methodology for diagnosing hard bugs and performance regressions by building tight feedback loops, reproducing, hypot

Provides utility functions for GSAP animations, including math operations, array handling, and value mapping.

Maintains a REGRESSION.md ledger of module dependencies and executable acceptance commands so that changes to one module are verified agains

Scaffolds, develops, installs, and verifies Paperclip plugins using a CLI-first local development workflow.

Performs a one-shot quality audit of an entire repository, ranking bugs, security holes, load risks, missing tests, slow paths, and code to

Reviews code changes for correctness, safety, performance, test coverage, and unnecessary code, explaining each finding in plain English.

Reviews a code diff like a senior engineer, reporting bugs, security risks, load problems, missing tests, speed issues, and code to delete w

Iteratively fixes and re-runs failing CI checks on a GitHub pull request until all checks pass or a specific blocker is identified.

Guides and best practices for developing with Remotion, including setup, rendering, and interactivity.

This skill sets up the repository configuration for engineering skills, including issue tracker, triage labels, and domain documentation lay

Systematically identify and resolve technical issues by investigating root causes before applying fixes.

Launches and verifies Paperclip instances from source, npm, or hosted stacks by driving real UI/CLI journeys and capturing evidence.

A framework for defining and running formal evaluations of Claude Code sessions to measure reliability and prevent regressions.

Performs evidence-backed impact analysis before high-risk code, data, or interface changes and reconciles the planned impact with the actual

Analyzes GitHub/GitLab/Perforce PRs/MRs/CLs for review comments, failed checks, and incomplete descriptions to prepare changes for submissio

Designing testable, modular codebases with deep interfaces and clear seams

Reviews code changes since a fixed point along two axes—adherence to coding standards and faithfulness to the originating spec—and reports b

Fetches up-to-date library and framework documentation via Context7 MCP so answers about APIs, setup, and code examples reflect current vers

Provides Playwright end-to-end testing patterns including Page Object Model structure, configuration, CI/CD integration, artifact management

Diagnoses and fixes slow PostgreSQL queries by reproducing baselines locally, proving performance and correctness improvements, and deliveri

Analyzes codebase architecture to identify and visualize refactoring opportunities

Migrate TypeScript test files from `as` type assertions to @total-typescript/shoehorn for type-safe partial test data.

A coding assistant mode that solves tasks with the smallest complete change and explains it in plain language.

Provides the simplest, most efficient coding solutions for tasks like writing, fixing, refactoring, and reviewing code.

Performs a full-repository quality audit covering bugs, security, load handling, missing tests, performance, and dead code, reporting findin

Displays a one-shot scoreboard of measured savings (lines of code, tokens, cost, time) from the Ponytail agentic coding benchmark.

Helps write clear pull request bodies with summaries, diagrams, before/after evidence, and merge risk analysis.

Builds quick throwaway prototypes to test UI designs or logic/state models, with minimal effort and no persistence.

Creates detailed maintainer-grade reports for pull requests, including deep code review, design analysis, and merge recommendations.

Automatically generate and update release changelogs and GitHub release notes from Git history.

Reverse-engineer compiled binaries, packages, firmware, and apps to explain features, compare versions, or guide reconstruction using shippe

Sets up a scheduled agent and routine that runs a code-maintenance skill on a repository and opens pull requests automatically.