superspecverify
Use after implementation is complete to verify that code matches specifications. Ensures every Requirement is implemented and every Scenario has a corresponding test.
Découvrez les skills Documentation, comparez leurs dépôts et consultez leurs instructions.
Use after implementation is complete to verify that code matches specifications. Ensures every Requirement is implemented and every Scenario has a corresponding test.
Fast-track from idea to implementation-ready plan in one session. Combines brainstorm (4 phases) + validate + plan into a single flow. Use for small-to-medium features that don't need separate review points.
Use after specs are validated to create TDD implementation plan. Each Scenario in Specs becomes a test case in the plan. Combines bite-sized TDD tasks with spec references.
Execute Gemini CLI for AI-powered code analysis and generation. Use when you need to leverage Google's Gemini models for complex reasoning tasks.
Use before any feature work. Progressive design flow with four phases: Explore → Propose → Design → Spec Outputs: proposal.md, design.md, specs/*.md
Execute Codex CLI for code analysis, refactoring, and automated code changes. Use when you need to delegate complex code tasks to Codex AI with file references (@syntax) and structured output.
Phase execution protocol to prevent context drift during long development sessions. MUST be read at the start of each phase and referenced in the last TODO of each phase. This ensures AI maintains context acros...
External AI code review using Codex or Gemini. Reads project.yaml to determine which provider to use for frontend vs backend tasks. Includes hallucination detection to avoid blindly applying AI suggestions.
Use to validate spec documents for correctness, completeness, and consistency. Ensures specs follow SuperSpec format and are ready for implementation.
Use when about to claim work is complete, fixed, or passing. REQUIRES running verification commands and confirming output before making any success claims. Evidence before assertions, always.
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies. Dispatch one agent per independent problem domain for parallel execution.
Use when starting feature work that needs isolation from current workspace. Creates isolated git worktrees with smart directory selection and safety verification.
Guide for UI styling with semantic design system classes. Never use gray-* colors or arbitrary Tailwind utilities.
Guide for validating external data with Zod. Use when fetching from APIs, processing webhooks, or handling user input.
Evaluate external PR suggestions (Gemini, CodeRabbit) against project conventions. Use when asked to review or agree with external code review comments.
Enforce TypeScript type organization rules. Use when creating types, interfaces, or DTOs. All types MUST go in /types directory - never inline in components.
Enforce the internal API proxy layer pattern. Use when adding API endpoints, fetching data from backend, implementing HMAC signing, or working with lib/api/* or app/api/*.
Guide for writing tests. Vitest for unit tests, Playwright for E2E. Includes SEO and sitemap testing patterns.
Enforce URL rules to prevent SEO issues and cost incidents. CRITICAL - never add searchParams to listing pages ($300 incident).
Guide for optimizing bundle size and images. Server Components by default, use image quality caps.
Guide for CSP, security headers, and external scripts. Use fetchWithHmac for APIs, safeFetch for external services.
Guide for internationalization with next-intl. Use when adding links, translations, or localized URLs. Always use Link from @i18n/routing.
MANDATORY checklist before writing ANY new code. Use this skill FIRST before implementing any feature, component, utility, type, or hook. Enforces DRY principle and existing pattern reuse.
Guide for adding environment variables. Use when adding new env vars to ensure all 4 locations are updated (code, SST, workflow, GitHub secrets).
Le classement utilise les étoiles du dépôt GitHub indexé. Plusieurs skills peuvent partager ce nombre : ce n’est ni un nombre d’installations ni une note de qualité.