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 →Master the GitHub CLI (gh) for complete Software Development Lifecycle management. Use this skill when managing GitHub repositories, issues, pull requests, GitHub Projects, releases, Actions workflows, secrets, Pages deployments, wikis, Codespaces, and automating via the GitHub API. Covers the entire SDLC from terminal without browser interaction.
>
Remove dead code, unused imports, stale files, and fix code quality issues found by /audit
Identifies and fixes ESLint errors and TypeScript type issues across the codebase.
Review, categorize, and approve unapproved YNAB transactions via DuckDB staging + YNAB MCP
1. Read `progress.json` to get current state
Owns the full FastAPI stack for web-cv-converter — routes, services, Pydantic schemas, Supabase migrations, Auth0 configuration. Performs OWASP self-review before every handoff.
Reviews React and TypeScript code for correctness, security, hook usage, and multi-tenancy safety. Use when reviewing generated React components, hooks, or TypeScript files, or when asked to do a code review on frontend code.
Workflow полного health check проекта. Architecture → Security → Quality → Report → (Fix). Use before major release or onboarding.
Use when producing a read-only CoordExp audit of code, configs, specs, artifacts, docs, progress notes, or OpenSpec changes for correctness, reproducibility, pipeline, and eval-validity risks.
Use when the user needs to design, implement, audit, or govern Storybook stories for frontend components, design systems, page states, interaction tests, accessibility checks, visual baselines, or component-driven UI workflows.
Codex entry skill for request-traceable implementation planning. Use when a request needs one or more executable plan artifacts under `./plans` after resolving blocking product policy, UX, contract, schema, validation, state, permission, and any upstream UI-direction ambiguity handled by `brainstorm` or `design-discovery`, with a controller-facing `plan.md` written in plain Korean that keeps the user's request items visible, groups work by concrete boundaries, exposes public contracts before execution order, and links per-phase technical detail files for later `plan-review` and `plan-materialize`, plus registry-backed review wiki core/pattern guidance.
Strict senior-engineer code review for code changes, PRs, diffs, generated code, architecture reviews, and bug hunts. Use when asked to review quality, maintainability, structure, tests, security, performance, error handling, or production readiness.
Audit and improve CSS transition declarations. Replace generic easing with intentional curves.
Coordinates planning, implementation, review, de-sloppification, and documentation for PythonExerciseGeneratorAndDistributor changes against ACTION_PLAN.md.
Run safety-first repository audits for OpenClaw Bakery with evidence-based findings and low-risk remediation. Use when the user asks to audit, clean up, review operational health, or align docs/config/paths without risky architectural rewrites.
Use when implementing, debugging, or reviewing code in an existing app and the user wants simple changes, reuse of existing components and design patterns, strong code quality, TDD-minded testing, and no over-engineering. Also use when deciding how much testing is needed, when validating UI changes with Agent Browser, or when simplifying a spec before coding.
Use when the user wants to convert Markdown to DOCX, make a Microsoft Word version of a Markdown file, style Word output with a reference template, fix DOCX callouts, or customize the bundled Word template for notes, reports, and review docs.
description: Run the project’s quality gates, map results to spec acceptance criteria, and produce verification_report.json with evidence and severities.
>-
Draft, review, and update Software Design Descriptions using an IEEE 1016-2009-inspired structure with explicit architecture/views/elements formalization and output structure validation. Use this whenever a user asks to write an SDD, assess SDD quality/completeness, align design docs to IEEE 1016 concepts, map PRD requirements to design, produce architecture/interface/data design sections, generate remediation-oriented gap reports, perform SDD review-only gap analysis, check whether an SDD has drifted from the codebase, or update an SDD after architecture changes.
Code review from a fresh context. Run in a separate session — no knowledge of implementation decisions.
Check Memeya's USDC reward wallet balance, view distribution history, and manually trigger reward payouts.
Reviews an existing AI agent (or agent design) against all 22 patterns from "Patterns for Building AI Agents" (Bhagwat & Gienow, 2025). Produces a scored checklist with specific recommendations for im
Handles git and GitHub operations using the gh CLI. Use when the user asks about pull requests (PRs), GitHub issues, repo management, branching, merging, rebasing, cherry-picking, merge conflict resolution, commit history cleanup, pre-commit hook debugging, GitHub Actions workflows, or releases. Covers creating and reviewing PRs, watching CI checks, interactive rebasing, branch cleanup, submodule management, and repository archaeology with git log/blame/bisect.
Autonomous optimization loop that improves any measurable thing. Point it at files to change, a command to check, and a number to improve — then walk away. Works with any AI agent. Use when the user wants to autonomously optimize code performance, ML training, Docker images, SQL queries, prompts, CI speed, bundle size, Kubernetes configs, or any domain with a measurable score. Triggers include requests like 'optimize this', 'improve performance', 'make this faster', 'reduce allocations', 'autoimprove', 'run the optimization loop', 'let it run overnight', or when the user has an improve.md file.
grill-me-coding
Address actionable GitHub pull request review feedback. Use when the user wants to inspect unresolved review threads, requested changes, or inline review comments on a PR, then implement selected fixes. Use the GitHub app for PR metadata and flat comment reads, and use the bundled GraphQL script via `gh` whenever thread-level state, resolution status, or inline review context matters.
Generate a code review with atomic views and remediation requirements for the current branch vs a base branch. Use this skill when the user wants a code review, branch review, diff review, change analysis, pre-merge review, or quality pass on their current work. Also use when the user asks to understand what changed on a branch, wants a narrative explanation of their changes, or needs a review document before merging.
Ensures .NET/C# code follows modern best practices, SOLID principles, and industry standards for production-ready applications
Finish a development branch — verify tests, merge/PR/cleanup.
Monitor pipeline agent health, task update buffers, and live agent state. Use when checking agent health, investigating task progress, querying pipeline agents for diagnostic state, or reviewing historical health trends. Covers query_agent_health, query_health_history, query_buffer, and query_pipeline_state skills.
Use when creating, reviewing, or migrating telemetry pipelines that collect, transform, route, buffer, validate, and deliver logs, metrics, traces, events, or profiles across tool-agnostic observability systems.
Verify academic paper references — automated ground truth check + live arXiv verification + multi-agent cross-review protocol.
Main Gatorbots FRC orchestrator policy for deterministic worker routing and honest answer modes.
Performs a final quality pass fixing alignment, spacing, consistency, and micro-detail issues before shipping. Use when the user mentions polish, finishing touches, pre-launch review, something looks off, or wants to go from good to great.
Tracks API spending across Anthropic, OpenAI, and Google keys with SQLite storage and daily Telegram digests
Define, update, and review design system components used across page specifications.
Top-level workflow loops that chain multiple personas autonomously. Use *fix, *review, *impl, *qa, or *plan sprint instead of manually invoking each persona in sequence.
Create and manage multi-agent build packets. Use when you need parallel agent lanes, separate handoff briefs, completion evidence, a master log, or an archive/review workflow for another project.
Parallel code review using both review-pr and codex-cli-review. Spawns two independent subagents, collects findings, outputs side-by-side reports with consolidated conclusion. Triggers: dual review, double review, full review, review with codex. Not for: single-tool review (use /review-pr or /codex-cli-review), pre-landing diff review (use /review).
Harden Carni-mvp CI with governance gates, secret scanning, TruffleHog-style checks, deterministic validation, and release discipline for protected delivery paths.
'Runs the expansion analyzer to identify gaps, missing capabilities, undocumented decisions, and improvement opportunities in the codebase. Produces a ranked list of suggestions with rationale, impact scores, and effort estimates. Never creates tasks in analysis mode — all suggestions require explicit human approval. Use --save to persist suggestions to disk for later review.'
Build and review Taiwan stock day-trading plans, pre-market checklists, intraday entry/exit rules, day-risk controls, and `quantitative-trading-decision-system` workflow checks. Use when the user asks about 台股當沖 planning, 盤前準備, 盤中補充標的, VWAP/量能/大盤條件, 停損停利, force-exit timing, or debugging the `當沖` project. Do not use for long-term investing, pure財報研究, or classic波段 setups unless the user explicitly wants an intraday adaptation.
Guide for creating well-structured conventional commits. Use when creating git commits, drafting commit messages, or reviewing commit message quality.
UI/UX design systems, Figma workflows, accessibility checklists, browser-based visual audit with automated screenshots, component specification patterns, and design-to-code handoff procedures. Use when designing interfaces, creating design systems, auditing accessibility, reviewing existing UI via browser screenshots, specifying components, or preparing design handoff for developers.
Development conventions and architecture guide for the OpenClaw CLI CLI repository.
Umbrella skill for building, reviewing, designing, porting, and migrating Avalonia applications with modern XAML/C# patterns on Avalonia 11.3.12. Use when the request is broad Avalonia work or when another specialist skill has not yet been selected; route quickly to focused skills for startup, bindings, styling, controls, layout, rendering, testing, design systems, or HTML/WinForms/WPF/WinUI/Avalonia 12 migration work.
High-rigor scholarship and PhD application preparation coach. Use when someone wants structured application prep, document review, narrative bank management, professor outreach coaching, LoR guidance, or application tracking. Supports quick prep (near deadline) and full-system coaching across STEM, humanities, social sciences, and interdisciplinary programs.
Evidence-based endurance coaching protocol (v11.25). Use when analyzing training data, reviewing sessions, generating pre/post-workout reports, planning workouts, answering training questions, or giving endurance coaching advice. Always read or fetch athlete JSON data before responding to any training question.