verify-changes
Use this skill after implementing code to verify correctness before notifying the user.
Découvrez les skills Bases de données, comparez leurs dépôts et consultez leurs instructions.
Use this skill after implementing code to verify correctness before notifying the user.
Create and maintain PLANS.md for long-running or risky tasks.
Update this documentation kit safely from upstream.
Use when the user asks for a plan or the task is complex/ambiguous. Enforces AGENTS.md workflow and encourages loading other relevant skills.
Maintain living architecture maps and interaction flows.
When explicitly authorized, execute end-to-end with a phased plan in one run until Definition of Done is met.
Detect repo tooling and patterns, then record a dynamic manifest.
Ask whether to proceed with refactoring after architecture mapping.
Document architectural decisions with context and consequences.
Confirm a requested feature change via AERA Debug mode (Debug view, console, logs). Use when validation must happen through Debug UI or when you need to add debug-only probes/commands to make a change observabl...
Ask clarifying questions and get approval before implementation.
Maintain Cupertino + Lumina visual consistency for the AERA frontend. Use when creating or modifying UI, CSS/tokens, layouts, components, menus/overlays, tables, charts, or interaction states in this repo.
Run the full autonomy bootstrap with minimal questions.
Enforce Definition of Done checks from QUALITY_GATES.md.
Detect incorrectly installed skills and report issues.
Validate changes against .agentignore before commit.
Install agent docs without overwriting or deleting existing content.
Extract recurring patterns from action logs and propose automations.
Compress long-run context into handoff, backlog, and action logs.
Analyze and document the impact of planned changes.
Detect legacy patterns and guide compatibility decisions.
Enforce test-first changes and debug based on failing tests.
Use this skill when adding new backend functionality (API endpoints + business logic).
Enforce .agentpolicy rules before executing commands.
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.