kevintsai1202

@kevintsai1202 source repo

34 skills
0 followers
8 hours ago last updated

▌
Add Skill Installer · kevintsai1202 bundle
安裝 Agent Skills 到本機。當用戶想要「安裝技能」、「新增 skill」、「從 GitHub 安裝 skill」、或「install skill from repo」時觸發此技能。

0
installs
▌
Buddy Reroll · kevintsai1202 bundle
幫助使用者刷出指定物種或稀有度的 Claude Code /buddy 寵物（五星/legendary）。當使用者想要重置 buddy、刷特定物種、刷 legendary 五星寵物時觸發。

0
installs
▌
Course Issue Publisher · kevintsai1202 bundle
Publish course entries from the local `course/` directory to the `kevintsai1202/course-schedule` GitHub repository and verify that they appear on the GitHub Pages site. Use this skill whenever the user says a course is already prepared under `course/<folder>/`, wants to publish one folder or all unpublished folders, wants image links inside markdown to work on the page, or wants to avoid re-uploading courses that were already published.

0
installs
▌
Exam System Project · kevintsai1202 bundle
Project-specific workflow for the Exam System repository. Use when working in this repo to follow its required order: read docs, update spec/api/todolist first, implement frontend or backend changes on Windows PowerShell, validate with the correct build/test commands, then commit, push, and prepare a release report.

0
installs
▌
Fda Document Guide · kevintsai1202 bundle
FDA medical device software certification document generation guide. Based on EN 62304, EN ISO 14971, and ISO/TR 24971 standards, helps generate 13 required regulatory documents (Documentation Level Evaluation, Software Risk Management Plan, Hazard Analysis, SRS, SDS, V&V testing documents, Risk Management File, etc.). Use this skill whenever the user mentions FDA/CE/MDR certification, software risk management, IEC 62304, ISO 14971, hazard analysis, software verification and validation (V&V), medical device software submissions, or asks about any regulatory standard requirements. Standard clauses are built into the references/ directory and can be referenced directly without any external connection.

0
installs
▌
Gemini Watermark Remove · kevintsai1202 bundle
移除 Google Gemini(Nano Banana)生成圖片中的浮水印，使用逆向 Alpha 混合演算法精確還原被覆蓋的像素

0
installs
▌
Skillsmp API · kevintsai1202 bundle
Search and discover AI skills from the SkillsMP marketplace

0
installs
▌
Webmcp Development Guide · kevintsai1202 bundle
Use when designing, implementing, reviewing, testing, or documenting browser-side WebMCP tools. Covers document.modelContext, Imperative and Declarative APIs, tool schemas, page state, API/MCP boundaries, security, and Inspector/Evals verification. It is not a replacement for a server-side MCP implementation guide.

0
installs
▌
Deep Memory · kevintsai1202 bundle
Use at the start of any task, whenever another skill is about to be invoked, or when a task looks complete — governs looking up and writing back this project's memory (knowledge-base/, experience/) so prior solutions and pitfalls get reused instead of rediscovered. Delegates semantic/hybrid retrieval to the chroma-hybrid-search sub-skill and GitHub backup/restore to the memory-backup sub-skill.

0
installs
▌
Memory Backup · kevintsai1202 bundle
Exports the local knowledge base (knowledge-base/, experience/) and ChromaDB vector data as JSONL, then backs it up to a dedicated private GitHub repository. Also supports cross-device restore from GitHub. Typically invoked by deep-memory after new knowledge-base or experience entries are written in a session, but can also be run directly by the user at any time to back up or restore.

0
installs
▌
Memory Import · kevintsai1202 bundle
Imports memory data from external systems — ChatGPT memory export, Claude Code local auto-memory files, or a legacy auto-skill project's knowledge-base/experience data — into this project's cold or hot store. Supports --dry-run preview and a safe merge that never overwrites without --force. Typically run once by the user right after a fresh deep-memory install if they have existing memory data to bring over (deep-memory's Step 0.5 mentions it), but can be run any time. Never scans the filesystem on its own — the user must point it at an explicit --input path.

0
installs
▌
Chroma Hybrid Search · kevintsai1202 bundle
Performs high-precision hybrid retrieval (BM25 + ChromaDB vector search + BGE-Reranker semantic re-ranking) over the local hot store (knowledge-base/, experience/) and cold store (cold-notes/raw.jsonl). Use this skill when high-accuracy code or solution retrieval is needed and AI hallucination must be minimized. Running for the first time in a directory will automatically trigger initialization. Typically invoked by deep-memory's knowledge-base retrieval step when keyword matching is insufficient, but can also be called directly for ad-hoc searches.

0
installs
▌
Legal Graph · kevintsai1202 bundle
引導 Agent 彙整案情事實、法條、判決、爭點、當事人、證據、契約義務關係（契約→條款→義務三層模型，含風險評級上色）與構成要件涵攝結果（法條→要件→事實該當性上色），產出與 law-powers 之 index.html/data.js 相容的 superset 法律關係圖譜資料，並寫入 data.js。另可依需求以自包含 SVG 產出「訴訟決策樹＋因果關係徑路圖」單檔雙面板圖（決策分支錨定條號判決、因果評價轉錄涵攝表），供書狀附圖與列印。

0
installs
▌
Legal Research · kevintsai1202 bundle
台灣法律問題的強制入口——凡涉及中華民國法規、判例、釋字、訴訟程序，或任何法律數值（裁判費、執行費、規費、罰鍰、賠償上限、法定期間、時效、級距門檻）的提問，一律先執行本技能。觸發語包含「要繳多少」「裁判費」「訴訟費用」「告人要花多少錢」「時效多久」「幾年內要提」「罰多少」「條文怎麼規定」。即使問題看似只需查一條條文、即使 taiwan-legal-db 與 dr-lawbot 工具已可直接呼叫，仍不得略過本技能自行檢索——本技能規範的授權子法反查與數值回歸驗算，是母法條文查詢本身看不出來的必要步驟（實測：僅查民訴 77-13 會答成裁判費 3000 元，實際應為 4100 元）。內容涵蓋查詢路由、雙軌並行檢索、引用驗證與信任閘門，並於完成後提供後續分析轉介選單。

0
installs
▌
Legal Brainstorming · kevintsai1202 bundle
當面臨法律爭議、案件諮詢或準備訴訟策略時，此技能用於釐清案情事實、梳理法律關係、規劃抗辯論點與整理證據清單，並於報告完成後主動提供後續分析選單（判例檢索、判決分析、要件涵攝、關係圖、合規審查、文字校訂）供使用者選擇轉介。

0
installs
▌
Legal Case Analysis · kevintsai1202 bundle
對經檢索驗證之判決進行深度分析：單一判決拆解（事實、爭點、兩造主張、法院論理、結論）、多判決見解歧異比較、實務見解演變時間軸，以及判決對本案之適用性評估（可援引／可區辨）。所有見解摘述必須錨定理由書原文段落，禁止憑記憶評析判決或杜撰裁判要旨。

0
installs
▌
Lawbank Query Builder · kevintsai1202 bundle
法源法律網（lawbank.com.tw）檢索式產生器——當問題涉及行政函釋、法學論著、修法沿革／立法理由、判解函釋交叉關聯，或 taiwan-legal-db 檢索空結果而使用者持有法源法律網會員時，本技能將需求轉譯為法源布林檢索式（＆且、＋或、－不含）與分區入口，由使用者本人登入人工檢索；並規範使用者貼回資料的驗證與標註紀律。觸發語包含「法源」「lawbank」「函釋」「行政函釋」「法學論著」「修法沿革」「立法理由」。本技能絕不自動連線法源、絕不索取或儲存帳號密碼。

0
installs
▌
Legal Element Analysis · kevintsai1202 bundle
對檢索取得之法條進行構成要件拆解與逐要件涵攝：民事請求權基礎檢驗與刑事三階層審查，輸出涵攝表、該當性結論與證據缺口清單。要件拆解須有條文原文或判決理由書依據，禁止自創要件或以推測補事實。

0
installs
▌
Compliance Verification · kevintsai1202 bundle
用於審查合約草稿與商業行為的適法性，比對民法、勞基法、個資法、公司法等法規，標示高、中、低風險條款並給出修改建議。

0
installs
▌
Legal Writing Humanizer · kevintsai1202 bundle
將一般中文、中國大陸法律用語或帶有 AI 寫作痕跡的法律文字，校訂為自然、精確且符合中華民國（台灣）法律實務的專業語體。使用者提到法律用語轉換、書狀潤稿、契約文字調整、法律意見去 AI 味、台灣法務語氣、存證信函或法律文件校訂時，務必使用本技能；不得藉改寫新增法律主張、法源或改變權利義務。

0
installs
▌
Official Document Drafting · kevintsai1202 bundle
用於撰擬或校訂中華民國行政機關公文書（函、書函、公告、簽等），依《文書處理手冊》及《政府文書格式參考規範》正確套用段落結構、用語與格式規則。當使用者要撰寫發文予政府機關之函、內部簽呈、公告或開會通知，或詢問公文用語、稱謂語、期望及目的語時，使用此技能。

0
installs
▌
Teaching Site · kevintsai1202 bundle
Use this skill as the main entry point whenever the user wants to build, plan, or evolve an interactive teaching website / course microsite / workshop landing page — from a blank slate, from existing course materials, or any state in between. Triggers on broad phrases like "做課程網站", "做教學網頁", "做工作坊網站", "把講義變網頁", "course microsite", "workshop site", "interactive lesson page", "multi-day curriculum website", "做一套課程". This skill is the router for the whole teaching-site pipeline (outline → content → SPA → interactions → visuals → corporate / ebook) and dispatches to the 10 specialised sub-skills as needed. Prefer this when the user's request is broad or unclear about which stage they're at — sub-skills (e.g. `course-ebook-publishing`) are still triggerable directly for stage-specific requests.

0
installs
▌
Web Content Audit · kevintsai1202
Use this skill when you need to verify cross-file data/asset consistency in a content-driven site — not "does it render?" (that's `web-visual-verification`) but "do the data, files, and references all line up?". Triggers on phrases like "盤點內容", "稽核資產", "對照 course-data 跟 markdown", "找缺圖", "task ID 有沒有重複", "quiz 編號 vs 硬編碼總數", "audit", "content audit", "content drift", "asset coverage", "three-way sync check", "cross-file consistency", "find missing illustrations", "資料一致性檢查", "找出該補插圖的地方", or any moment when the user senses divergence between source files and deployed data. Output is a human-readable markdown report, not a pass/fail. Pair with `web-visual-verification` for full pre-release coverage.

0
installs
▌
Web Visual Assets · kevintsai1202
Use this skill when a content-complete website has missing/placeholder images and needs visual assets — scenario illustrations, tool screenshots, instructor cards, conceptual diagrams, classroom location maps, QR codes. Triggers on phrases like "插圖", "工具截圖", "QR", "講師卡", "地圖", "示意圖", "Playwright 爬蟲", "AI 生圖", "visual assets", "screenshots", "illustrations", "QR codes", "instructor cards". This skill covers the four asset sources (scraping, AI generation, hand-drawn SVG, generated codes), the PNG-first + SVG-fallback render pattern, and verification scripts. Usually invoked AFTER interactions are wired (so missing images are visible), but can be invoked earlier if assets are pre-planned.

0
installs
▌
Course Outline Design · kevintsai1202
Use this skill when starting a new course / workshop / training program and you need to design its outline before any lecture content is written. Triggers on phrases like "規劃課程", "課程大綱", "幾天怎麼排", "學習目標", "workshop outline", "syllabus design", "course outline", "learning objectives", "課程總覽", or when the user has a topic and audience but no structure yet. This skill produces the skeleton (.md files) that all downstream stages — content authoring, SPA conversion, visual assets — depend on. Always invoke this BEFORE jumping into writing lecture notes or building a site.

0
installs
▌
Static Spa Conversion · kevintsai1202 bundle
Use this skill when you have structured course content (or any chapter-based dataset) in markdown form and need to turn it into a working interactive website — without picking a framework, without a build step. Triggers on phrases like "做成網頁", "轉成 SPA", "course-data.js", "render 函式", "把講義變網頁", "static site from markdown", "vanilla JS site", "no-framework site", "single-page app from markdown". The output is a vanilla HTML + JS single-page app that opens with `npx serve` and persists state in localStorage. Always invoke AFTER `course-content-authoring` (content stable), BEFORE `static-spa-interactions` (this skill produces the scaffold; interactions adorn it).

0
installs
▌
Course Ebook Publishing · kevintsai1202
Use this skill ONLY AFTER a teaching site is feature-complete and stable, when the user wants to produce a printable / archivable / shareable version — typically a PDF ebook with cover, page numbers, table of contents, or a DOCX for editorial review. Triggers on phrases like "做電子書", "產 PDF", "ebook", "PDF ebook", "印給學員", "課程講義 PDF", "course handbook", "DOCX deliverable", "book-style layout", "build:ebook". Do NOT invoke during early site development — this is a downstream consumer that breaks if the site changes underneath. The skill produces PDF via Playwright `page.pdf()` and DOCX via pandoc, both from a single composed markdown source.

0
installs
▌
Static Spa Interactions · kevintsai1202
Use this skill when a vanilla static SPA scaffold exists and needs to feel like a polished product — progress persistence, sidebar + scrollspy navigation, dark/light theme, content zoom, responsive (mobile sidebar overlay), keyboard nav, fade-in entrances, toast notifications, iframe modal viewers, quiz scoring with section back-links. Triggers on phrases like "加進度勾選", "響應式", "暗色模式", "縮放", "scrollspy", "sidebar 收合", "手機版", "RWD", "dark mode", "progress tracking", "quiz UX", "interactive polish". Always invoke AFTER `static-spa-conversion` (renders working), as a standalone enhancement layer.

0
installs
▌
Web Visual Verification · kevintsai1202
Use this skill whenever you need to verify that a web UI actually works the way it's supposed to — clicking through flows, asserting state, catching console errors, taking screenshots across multiple viewports for visual review. Triggers on phrases like "驗證網頁", "verify the site", "Playwright tests", "visual regression", "RWD verification", "screenshot comparison", "responsive check", "console error check", "看看手機版有沒有壞", "視覺驗證", "Playwright 測試", "截圖比對", "RWD 驗證", "驗 sidebar", "看 console 有沒有錯", or any post-change moment where the user wants to know "did I break anything?". This is the runtime-behaviour verifier — use `web-content-audit` instead if the question is about file/data consistency rather than rendered behaviour.

0
installs
▌
Course Content Authoring · kevintsai1202
Use this skill when an outline exists and you need to fill in the actual teaching content — lecture notes, sample materials (CSV / YAML / MD samples), quiz items, and any course-specific artifacts like prompt templates or worksheets. Triggers on phrases like "寫講義", "補素材", "出測驗題", "提示詞範本", "撰寫課程內容", "lecture notes", "course material", "quiz questions", "quiz authoring", "prompt templates". Always invoke AFTER `course-outline-design` is stable, BEFORE `static-spa-conversion`. This skill produces only the raw `.md`/`.csv`/`.yaml` artifacts — turning them into a website is the next stage's job.

0
installs
▌
Course Corporate Edition · kevintsai1202
Use this skill when an existing multi-day public-class teaching site needs to be condensed and re-packaged for a corporate in-house training — typically a half-day or one-day intensive (e.g. 4-day public class → 6-hour corporate). Triggers on phrases like "企業包班", "濃縮版", "客製化課程", "corporate edition", "intensive version", "in-house training", "single-file deliverable", "客戶端課程", "壓縮成一天", or when the user wants a single-folder zippable deliverable for a client. This skill builds on top of an existing site — picking units, condensing time allocation, embedding data into a single HTML, setting up asset fallback chains, and producing an offline-ready package.

0
installs
▌
Teaching Site Design System · kevintsai1202 bundle
Use this skill whenever you're making visual / styling decisions on a teaching site or content-driven SPA — picking colors, fonts, badge styles, card layouts, hero treatments, dark mode tokens, or any "how should this look?" question. Triggers on phrases like "視覺風格", "設計系統", "色票", "字體", "暗色模式", "玻璃卡片", "提示詞徽章", "Day hero 怎麼設計", "design tokens", "design system", "color system", "glass cards", "typography", "look and feel", "視覺一致性", or any moment when the user evaluates aesthetics rather than behaviour. This is the cross-cutting visual authority — every stage (SPA, interactions, corporate edition, ebook) reads tokens from here so the brand stays coherent across formats.

0
installs
▌
Humanizer Zh Tw · kevintsai1202 bundle
去除文字中的 AI 生成痕跡。適用於編輯或審閱文字，使其聽起來更自然、更像人類書寫。基於維基百科的「AI 寫作特徵」綜合指南。偵測並修復以下模式：誇大的象徵意義、宣傳性語言、以 -ing 結尾的膚淺分析、模糊的歸因、破折號過度使用、三段式法則、 AI 詞彙、否定式排比、過多的連接性短語。若使用者要求清理不可見 Unicode、文字型 AI 浮水印或檔案 provenance，改用 $text-watermark-cleaner-zh-tw，不要在本 skill 中把語氣改寫冒充成浮水印已移除。

1
install
▌
Text Watermark Cleaner Zh Tw · kevintsai1202 bundle
清理使用者擁有或獲授權文字中的不可見 Unicode、zero-width、tag characters、異形空白與文字型 AI provenance/watermark，並提供統計式改寫的 best-effort 流程。適用於繁體中文、簡體中文、英文、Markdown、HTML 與純文字；當使用者說「去文字浮水印」、「去 Claude 浮水印」、「清理不可見字元」、「清除 AI provenance」或要求檢查文字標記時使用。文字浮水印與檔案 C2PA/EXIF metadata、可見圖片浮水印是不同問題，後兩者不由本 skill 直接處理。

1
install

34 published skills