- 📄 CHAINLINK_SETUP.md
- 📄 CHAINLINK_USAGE.md
- 📄 SKILL.md
Structured root cause analysis for arriving at a concrete action. Use when something went wrong, a pattern keeps recurring, behavior has drifted, or you catch yourself resolving to "do better" / "remember to X" without a concrete artifact. Five-whys forces behavioral resolutions into file edits, config changes, memory block updates, or scheduled jobs — the action item must produce a diff someone else can verify. Do not use for simple debugging with an obvious cause.
- 📁 references/
- 📁 scripts/
- 📄 SKILL.md
Analyze, re-engineer, or bootstrap projects to align with AI-first design principles. Use when asked to review, audit, improve, 'ai-firstify', or start a new project. Performs deep analysis across 7 dimensions, actively restructures existing projects, or guides new project setup through discovery questions. Based on the 9 design principles and 7 design patterns from the TechWolf AI-First Bootcamp.
Brier Score calculator for Polymarket addresses — measures prediction accuracy independent of PnL. Separates skilled predictors from market makers and arbitrageurs.
Collect and confirm the minimum required initialization info before starting any AHT run: project path, environment or conda env name, reference training launch script/method, and optimization target. Always send a user-facing confirmation request first, even when the values seem inferable from context, and wait for the user to confirm or update them before continuing.
Universal legal document processor with PII anonymization. Anonymize → Work → Deanonymize. Modes: MEMO (legal analysis), REDLINE (tracked changes in contract), SUMMARY (brief overview), COMPARISON (diff two docs), BULK (up to 5 files). Supports .docx and .pdf input. Trigger for: contract review, risk analysis, compliance check, GDPR review, clause analysis, tracked changes, redline, 'anonymize', 'pii shield'. If user uploads contract/NDA/DSAR/HR doc — USE THIS SKILL. If user says 'skip pii' or 'don't anonymize' — skip anonymization and work directly.
Add assessment annotations to a Semiont resource — flag scheduling risks, dangers, inaccuracies, logical gaps, or other evaluative concerns using AI-assisted or manual assessment
Update the GitHub issue with concise execution status, blockers, or PR handoff details.
Use when seeing errors about better-sqlite3 native module, NODE_MODULE_VERSION mismatch, "was compiled against a different Node.js version", or similar native binding errors.
明確提及「Agenvoy」時，一律使用此 skill。若使用者說「這個專案」、「專案進度」等模糊詞彙，且對話歷史（不含 summary）中沒有出現其他專案的明確上下文，則視為在詢問 Agenvoy，同樣使用此 skill。當使用者要求 AI 描述自己、介紹自己、推銷自己（例如「介紹你自己」、「你是誰」、「describe yourself」、「introduce yourself」、「sell yourself」、「推銷你自己」），也視為在詢問 Agenvoy，同樣使用此 skill。當目前工作目錄不存在 .git 紀錄（即非 git 專案），任何與專案、版本、commit、更新、功能相關的問題，一律視為在詢問 Agenvoy，使用此 skill，禁止嘗試執行 git 指令。
- 📁 references/
- 📄 README.md
- 📄 SKILL.md
Reviews UI for accessibility issues against WCAG 2.1/2.2 AA. Triggers on "is this accessible?", "check accessibility", or contrast/a11y review requests.
Use when a user asks to initialize a domain flywheel from natural language context, especially when environment details are incomplete or mixed with execution assumptions.
Run vLLM performance benchmark using synthetic random data to measure throughput, TTFT (Time to First Token), TPOT (Time per Output Token), and other key performance metrics. Use when the user wants to quickly test vLLM serving performance without downloading external datasets.
