Skills（SKILL.md）は、AIエージェント（Claude Code、Cursor、Codexなど）に特定の能力を追加するための設定ファイルです。
詳しく見る →Reviews UI foundation, UI generation, and frontend implementation outputs or produced frontend code for consistency, usability, accessibility, responsiveness, correctness, and implementation readiness, then produces a structured QA verdict with required revisions.
Use this skill when building, reviewing, or refactoring Go code in this repository. It adds process discipline for TDD/ATDD, package boundaries, dependency injection, and review quality, and complements the always-on rules in AGENTS.md.
Session wrap-up. Update memories, check plans, review git state, check inbox, flag loose ends. Use before closing a session or compacting context.
Update branch memory files after completing work. Saves session history, key learnings, and collaboration observations to .trinity/ files.
Use when asked to prepare, review, or create a git commit for this repository. Follow the local commit policy, keep commits scoped to the intended changes, run the fast required checks, and write Conventional Commits that work with release-please.
Primary router skill for broad .NET work. Classify the repo by app model and cross-cutting concern first, then switch to the narrowest matching .NET skill instead of staying at a generic layer.
Create a git commit following project conventions. Use when asked to commit, create a commit, or finalize changes.
Add or update automated tests for a change using the repository’s verification rules in `AGENTS.md`. Use when implementing a feature, bugfix, refactor, or regression test; prefer stable integration/API/UI coverage and pull deeper test strategy from the bundled references.
Use for PR/code reviews and any task that benefits from a dedicated tmux sub-agent with per-task git worktrees; default path for reviewing diffs (read diff → summarize → run checks/tests) with automated monitoring.
Generate 5 distinct, production-grade UI variations for a requested new or existing section, component, page, flow, or shell, then help the user preview and apply one. Use when the user asks to add or redesign UI like a hero, pricing, navbar, auth flow, dashboard shell, ecommerce surface, or data component.
Use to update work item state on the project board at lifecycle phase transitions. Automatically invoked by /ai-brainstorm (refinement, ready), /ai-dispatch (in_progress), and /ai-pr (in_review). Also invoke manually for 'move this issue to in-review', 'update the board', 'mark as in progress', 'sync the work item state'. Fail-open: never blocks the calling workflow.
Test Architect & Quality Advisor (Quinn). Use for comprehensive test architecture review, quality gate decisions, and code improvement. Provides thorough analysis including requ...
Review a change set for correctness, regressions, security risks, and test coverage gaps.
Pipeline orchestrator that works Ada-labelled Planka cards through the full dev pipeline (story → plan → code → review). Use for autonomous development sessions.
Documentation creation and modification rules (docs/en/developer/ design documents, abstraction principles, structure, docs-first workflow). Use when creating documentation, updating design documents, writing architecture docs, checking which docs must change for a code change, asking about documentation structure/design principles, or working with files in docs/en/developer/ (fallback: docs/ja/).
Generate or update backend CRUD slices for SeeSpec ABP/.NET 8 features. Use when adding or changing a domain entity and its full backend path: Core entity, EntityFrameworkCore registration, DTOs, application service, service interface, permission hooks, migrations checklist, and related documentation/tests.
Use when the user pastes or describes an inbound advisor-agent email dispatched to this steward. Advisor emails end with `@lessersoul.ai` and carry a provenance signature. This skill verifies the brief's origin, extracts the request cleanly, and surfaces it to Aron for explicit review before any action is taken. Advisor-dispatched work never executes autonomously.
Reviews code for quality, security, and maintainability. Use after implementing features, completing significant work, or when user asks for code review.
Check whether implemented code follows the copied `references/rule/**` rule documents for module structure, dependency direction, naming, and Spring Boot/JPA/RabbitMQ/Liquibase conventions. Use when reviewing a diff, module, package, `build.gradle.kts`, or implementation for compliance with the copied rule references.
Execute tasks from the plan, track progress, and enable session continuation.
Review current changes against requirements, correctness, maintainability, and performance risk.
Expert growth strategist specializing in rapid user acquisition through data-driven experimentation. Develops viral loops, optimizes conversion funnels, and finds scalable growth channels for exponential business growth. Use when Codex needs this specialist perspective, workflow, or review style for related tasks in the current project.
failed, blocked, review follow-up, 추가 요구사항이 들어온 task를 reopen해서 repair loop를 재개한다.
Review and rate code implementations against issue cards and ExecPlans using git diffs, AGENTS.md, and PLANS.md. Use when a user asks for an implementation review, compliance check, regression risk assessment, or to validate changes against one or more issue cards or ExecPlans.
Generate a safe, step-by-step refactor plan minimizing risk and preserving functionality.
Create or update a feature spec under `docs/Features/` with business rules, user flows, system behaviour, verification, and Definition of Done. Use when the user asks for a feature spec, executable requirements, acceptance criteria, behaviour documentation, or a pre-implementation plan for non-trivial behaviour changes.
- Ensure the PR is conflict-free with `origin/main`.
Review, write, or fix unit tests in TypeScript repos that use Vitest, Angular TestBed, and component-driven rendering. Use this when adding tests, debugging flaky or failing tests, preventing ExpressionChangedAfterItHasBeenCheckedError, improving beforeEach setup, or standardizing safe render patterns. Do not use for end-to-end tests, backend integration tests, or broad refactors unrelated to test behavior.
Teach and apply Effect-TS practices (effect, @effect/schema, @effect/platform) for architecture, typed errors, Layers, dependency injection, resource safety, testing, and migration from async/await/Promise. Use when a user asks how to use Effect, requests best practices, wants to refactor to Effect, or needs mapping from platform modules to Node/Bun/Browser APIs.
Use when designing safe public APIs that wrap unsafe Rust code, adding unsafe blocks to existing types, reviewing unsafe code for soundness, or creating new types backed by raw pointers, MaybeUninit, or FFI
Use when implementing code changes to ensure quality, consistency, and maintainability.
Review checkpoint specs and tests to identify tests that encode ambiguous interpretations rather than explicit requirements. Use when asked to check checkpoint_N.md against test_checkpoint_N.py, when auditing tests for ambiguity, or when reviewing snapshot eval failures for interpretive issues.
Mark a draft pull request as ready for review.
Enforce the phase implementation workflow for Palette Kit v0.3 phases. Use when implementing a roadmap phase, creating a phase branch, validating the lint/typecheck/test/build pipeline, generating a patch for review, or preparing commit/push/PR steps.
Security review checklist for code and infrastructure
Manage current feature workflow - start, review, test, explain, or complete
Use when the user wants a guided review of a branch or diff in execution order through the changed behavior, with light workflow context only where needed.
Adversarially review proposed mew self-improvement loops for product-goal drift, M5 safety boundaries, weak evidence, missing verification, hidden rescue edits, and whether the loop should be approved, rejected, or revised.
Use when you have made code changes and want another CLI agent to review them before committing or continuing. Covers crafting review prompts, scoping reviewer permissions to read-only, interpreting feedback, and follow-up clarification sessions.
Pause for review every N tasks - selective autonomy pattern
Keep git history clean for each goal-oriented work unit. Use when Codex starts new implementation work, prepares a PR/review request, or needs to ensure unrelated commits/diffs are not mixed.
Use when developing, reviewing, or closing work in beamHO-bench that is governed by an active pending SDD, especially when the task requires reading PROJECT_CONSTRAINTS, active pending status, validation rules, and lifecycle-sync docs before editing code.
Use this skill when creating or refactoring Svelte 5 / SvelteKit frontend code, especially UI components, local feature UIs, component APIs, and Tailwind-based styling. This skill improves code quality by enforcing modern Svelte 5 conventions, strong component boundaries, clear state ownership, reusable modular logic, and disciplined Tailwind usage. Not intended for visual design exploration or branding work.
Generate a concise hotspot briefing from the wy-hotspot-monitoring workspace state. Use this skill whenever the user asks for a hotspot digest, trend summary, recent signal review, shift report, or operator-ready briefing, especially when they want a quick readout instead of raw JSON or the full dashboard.
Deploy any project to Vercel. **Always deploy as preview** (not production) unless the user explicitly asks for production.