GitHub Copilot skills: 740 worth installing

Refactors working code so it is easier to read, maintain and extend, without changing its behaviour.

Interviews you about your quality bar, writes it down as a contract with default thresholds, and stops agents from quietly lowering it.

Sets up rules files and project context for an agent session, and repairs the context when output quality starts to degrade.

Walks the agent through root-cause debugging when tests fail, builds break or behaviour stops matching expectations.

Plans the removal of old systems, APIs and features, and the migration of users or production database schemas to their replacements.

Records architecture decisions as ADRs and keeps documentation current when public APIs change or features ship.

Puts every non-trivial decision through an adversarial review in a fresh context before it stands, to expose hidden failure modes.

Builds accessible, responsive production UIs: components, layouts, state management and WCAG requirements.

Structures git work: atomic commits, branching, conflict resolution, cleaning up a messy working tree and pull requests.

Turns a vague idea into a sharp, actionable concept through structured divergent and convergent thinking, testing assumptions before you plan.

Delivers a change in thin slices that can each be verified, for work that touches several files or rolls out behind a feature flag.

Interviews you one question at a time until the agent is about 95% sure what you actually want from an underspecified request.

Adds logging, metrics, tracing and alerting so that production behaviour can be seen and diagnosed.

Finds and fixes performance problems across frontend, backend and database, including Core Web Vitals, load times and N+1 queries.

Breaks a spec or set of requirements into ordered, implementable tasks, with scope estimates and work that can run in parallel.

Audits and hardens code that handles user input, authentication, data storage or external integrations against common vulnerabilities.

Prepares a production launch with a pre-launch checklist, monitoring setup and a staged rollout plan.

Checks every implementation decision against official documentation, so the code cites its sources and avoids outdated patterns.

Writes a spec or PRD with objectives and scope before any code, for new projects, features and significant changes.

Drives implementation with the red-green-refactor loop: write a failing test, make it pass, then clean up.

Generates brand-kit images: brand guideline boards, logo systems and identity decks in several visual directions.

Produces raw, mechanical interfaces that mix Swiss typographic print with terminal aesthetics, for dashboards and editorial sites.

A GPT-tuned variant of the taste skill, with enforced layout variance, editorial typography and GSAP scroll motion.

Generates mobile app screen concepts and flows as images, for iOS, Android and cross-platform products.