Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Daily maintainer routine — open issues / PR review / build health / broken-link audit via canonical MAINTAINER-PIPELINE. TRIGGER when: user says "maintainer", "跑 maintainer", "daily PR review", "issue triage", or routine `twmd-maintainer-daily` fires.
Review a Taiwan.md PR via canonical MAINTAINER-PIPELINE. TRIGGER when: user says "審 PR", "Review PR", "PR #N", "免疫巡邏", "跑 PR review".
This skill should be used when the user asks to check system health, verify their setup is working correctly, or wants an overall status report. Trigger phrases include "check my system", "is everything working", "system status", "health check", "diagnose my setup", "what's the status".
This skill should be used when the user reports that their AI Trader is not triggering, not executing trades, or behaving unexpectedly. Trigger phrases include "why isn't my trader working", "strategy not executing", "no trades happening", "trader not triggering", "why no signals".
Use when checking the overall health of a skills library. Run doctor, validate, check for stale skills, and verify generated docs are in sync.
Use when evaluating whether a skill belongs in a library. Preview content, check frontmatter, validate structure, and decide whether to keep, curate, or remove.
Use when the agent has acquired a tool or instrument and needs to verify it is operational before first use in a critical task step. This skill performs a lightweight pre-use check via "focus on [TOOL] in inventory" and confirms readiness based on the system's response, ensuring the tool is functional before measurement, activation, or connection operations.
外部から GitHub PR が来たときの triage → 分類 → 対応判断 → (承認後の)応答/取り込みを 定型化する skill。他人のコードは checkout もスクリプト実行もせず gh 経由の remote diff だけで triage を完結させ、merge / close / comment などの外向き操作は必ず sotarok の 承認を経る。supply chain 攻撃(lockfile poisoning / postinstall / workflow 改変)の 予防を最優先に置く。 キーワード: PR, pull request, 外部コントリビュート, triage, スパム, supply chain, i18n, 翻訳
実機ブラウザでの有限探索 QA skill (Bounded DevTools Charter Runner)。クリティカル フロー表 (`.claude/agents/feature-planner.md:84-90`) を 9 チャーターに展開し、 操作数/リトライ/時間の上限つきで chrome-devtools MCP (優先) / claude-in-chrome (フォールバック) を駆動する。`packages/e2e` (決定的回帰) を置き換えず、視覚的・ 状態依存の確認と `--prod-build` (standalone ビルド) スモークを担当。findings は fix-or-drop、git push はしない。 キーワード: qa, QA, 探索的テスト, exploratory testing, devtools, chrome-devtools, claude-in-chrome, charter, prod-build, スモーク, collab QA
Codex を敵対的レビュアーに使い、指定スコープ (worktree 差分 / main 直作業範囲 / エリア) を review → 報告 (+ main clean なら低リスクだけその場修正) する単発レビュー skill。findings は事前の全件検証をせず unverified タグ付きで報告し、修正を適用する ものだけ着手時に実コードで裏取りする (verification-on-action)。PR/merge の自動化は しない (crowi は main-direct + integrate-worktree のため)。Codex が落ちたら Claude subagent で代替。 キーワード: review, code-review, codex, adversarial, レビュー, 品質