chienchuanw

@chienchuanw source repo

38 skills
0 followers
6 hours ago last updated

▌
Bpm · chienchuanw bundle
Estimate the BPM of a show's audio files and write each tempo into its song macro's `SetVar $songbpm` line on grandMA2, so the speed master tracks per song. Use whenever the user points at a folder of performance audio and wants BPM detected and pushed into the song macros — "analyze the BPM", "calculate tempos and update the macros", "fill in $songbpm from these tracks", "get the BPM for the set". Requires the gma2 MCP connected and the song macros already built (see the connect and setlist skills). Needs ffmpeg; pulls librosa via uv.

0
installs
▌
Gh Pr · chienchuanw bundle
Push a development branch and create or update a pull request via gh CLI using type-specific templates. Optionally updates README, docs, and project tracking before opening the PR. Use after finishing development, when ready to open a PR, submit work for review, or push changes.

0
installs
▌
Gh Dev · chienchuanw bundle
Create a branch linked to a GitHub issue via gh issue develop, then implement the work with regular convention-following commits. Use when starting work on an issue, picking up a ticket, developing a feature branch, or any intent to begin coding against a GitHub issue.

0
installs
▌
Gh Fix · chienchuanw
Read unresolved review comments on a GitHub PR, deeply evaluate each suggestion, then either fix it (via gh-dev + simplify) and push the result, or push back via gh-comment with a reasoned reply. Use when the user wants to address PR feedback, respond to reviewer comments, fix code-review suggestions, action a review, or any phrasing like "address the comments on PR #N", "respond to the review", "fix what the reviewer asked for", "go through the PR feedback", or "handle the review comments".

0
installs
▌
Feature · chienchuanw
End-to-end feature delivery workflow that orchestrates issue creation, branch setup, strict TDD implementation, PR opening, review-feedback handling, and session archival. Use when the user wants to ship a feature from start to finish with one command — e.g. "/feature 42", "/feature add RSS feed", "build this end-to-end", "deliver issue 17", "take this from issue to merged PR". Bundles the plan → TDD → PR → review-fix → archive arc that would otherwise require invoking 5+ separate skills.

0
installs
▌
Readme · chienchuanw bundle
Write a new README.md or improve an existing one. Use when the user asks to create, update, or improve a README, or wants to document a project and no other documentation skill is more specific.

0
installs
▌
Design · chienchuanw bundle
Act as a professional concert lighting designer on grandMA2: design each song's looks from its audio energy, its section map and the rig's real 3D geometry (MVR), then program them into that song's cues as preset-referenced values with layered fade times, per-song effects locked to the BPM speed master, plus a follow-spot sheet for the operator. Use whenever the user wants the lighting itself designed or programmed — "design the lighting for this song", "program the looks", "fill in the cues", "what should this chorus look like", "設計燈光", "編燈", "打光". Requires the gma2 MCP connected (connect skill) and the song's sequence with labelled empty cues already built (setlist + cuelist skills); the presets skill supplies the palettes this skill references. NOT for UI, web or graphic design (that is the separate design-canvas skill), and NOT for building the empty cue skeleton from a CSV (that is cuelist).

0
installs
▌
Gh Issue · chienchuanw bundle
Create structured GitHub issues via gh CLI using type-specific templates (bug, feat, refactor, doc, perf, security). Use when filing bugs, requesting features, tracking work, opening tickets, or any intent to create a GitHub issue — even implicit phrases like "track this" or "file this".

0
installs
▌
Connect · chienchuanw
Install, register, and verify the gma2 MCP server so Claude can drive a grandMA2 lighting console (or onPC) over Telnet. Use this whenever the user wants to "connect to grandMA2", "set up the MA2 console", "install gma2 mcp", "hook up the lighting desk", asks why the gma2 tools aren't available, or is about to run any other gma2 skill (presets / setlist / bpm) and the server isn't connected yet. This is the entry point for the whole gma2 plugin — run it first.

0
installs
▌
Cuelist · chienchuanw bundle
Import a song's cue-list CSV (section markers like PGM IN / INTRO / VER 1 / CHOR 1 with a number, name, and timecode) into that song's grandMA2 sequence, building each section as a labelled cue so the operator gets a navigable skeleton to program against. Use whenever the user hands over a per-song CSV of cues/sections and wants it pushed onto the console — "save this cue list to the sequence", "import these cues", "build the cues from this CSV", "turn this section map into cues for the song", or points at a file named like `S1_<song>.csv`. Assumes the song's sequence already exists (built by the setlist skill) and the gma2 MCP is connected (connect skill). Prefer this over hand-typing Store commands — it parses the CSV, maps it to the right sequence, guards against clobbering existing cues, and verifies the result.

0
installs
▌
Presets · chienchuanw bundle
Build grandMA2 Gobo / Color / Beam / Focus preset palettes from fixture-type XML profiles, programming the correct per-fixture-type DMX value for each named function so one logical preset works across mixed fixtures. Use whenever the user wants to "build presets", "make a color palette", "set up gobo/beam/focus presets", "program a preset pool", or hands over fixture XML files and a console to populate presets on. Requires the gma2 MCP server connected (run the connect skill first). Prefer this over hand-writing preset commands — it resolves fixture-specific values from the profiles instead of guessing.

0
installs
▌
Setlist · chienchuanw bundle
Build a grandMA2 per-song show system from a set list / rundown: for each song a selector macro, a sequence (with a home cue), a same-named executor on the "Set List" page, a per-song page and view, plus one cue on a master "Set List" cuelist coloured by artist. Everything is linked BY NAME so the existing "Song Change" macro engine drives it. Use whenever the user hands over a rundown / set list (PDF or text) and wants the song macros / sequences / pages / executors built or rebuilt, says "build the set-list system", "set up the songs", "program the show structure from this rundown", or wants to understand/map an existing song-macro engine. Requires the gma2 MCP connected (run the connect skill first).

0
installs
▌
SEO Meta · chienchuanw
Generates SEO metadata (slug, subtitle, tags, title, description) for markdown articles as YAML frontmatter. Use when the user asks to add metadata, SEO info, tags, or frontmatter to an article, or wants to prepare a markdown post for publishing.

0
installs
▌
Commit Msg · chienchuanw
Suggests 3 commit messages based on git diff and project conventions. Use when the user asks for commit message ideas, wants help writing a commit message, or has changes ready to commit.

0
installs
▌
Gh Archive · chienchuanw
Capture the current state of a project by updating documentation at the end of a development session. Invokes the readme skill to refresh README.md and the planning-with-files skill to record project status in a doc/ or docs/ directory. Use after merging a PR, finishing a feature, wrapping up a coding session, or any time the user wants to snapshot where the project stands — even implicit phrases like "archive this", "wrap up the session", "document current status", "update project docs", "snapshot this repo", or "I'm done for today". Also use when the user says "end of session" or asks to prepare documentation before stepping away from a project.

0
installs
▌
Gh Comment · chienchuanw bundle
Post formatted comments on GitHub PRs or issues, approve PRs, and merge PRs via gh CLI using purpose-specific templates (feedback, question, status, approval, request-changes). Use when commenting on a PR, leaving feedback, asking a question on an issue, posting a status update, approving a pull request, merging a PR, or any intent to interact with an existing PR or issue — even implicit phrases like "tell them", "approve it", "merge this", "LGTM", or "ship it".

0
installs
▌
End Of Day · chienchuanw
Run the user's bedtime wind-down — a pure recovery ritual to help sleep, reduce stress, and lightly consolidate the day's learning. It is deliberately the LAST thing of the day, run AFTER any accountability work: it NEVER touches todos, scoring, objectives, or 記帳 (those belong to daily-reviewer). A single-pass guided flow — the user answers one compact prompt block in one go — then the skill closes open mental loops, guides one short breath, appends one block to the month's wind-down log, and hands the user off the screen ("關螢幕、去睡") without asking anything further. Use whenever the user wants to wind down, decompress, or close the day for sleep — "end of day", "wind down", "我要睡了", "準備睡了", "睡前收尾", "收心", "幫我收個尾", "今天到此為止", "let me wind down before bed" — even if they don't name a skill. This is the bedtime bookend that comes AFTER daily-reviewer's retrospective: prefer daily-reviewer when the user wants to review, score, or 記帳 the day; prefer THIS skill when the user wants to relax, calm down, and go to sleep.

0
installs
▌
Health Audit · chienchuanw
Manually-triggered comprehensive codebase health audit covering security, architecture, documentation drift, dead code, and test coverage gaps. Produces a ranked markdown report, files prioritized GitHub issues, and proposes (but never autonomously opens) auto-fix PRs for trivial safe changes. Use when the user wants to audit, scan, review the health of, do a security check on, find dead code in, check doc drift in, or get a backlog of improvement issues for a repository — e.g. "/health-audit", "audit this repo", "do a full health check", "scan for issues to file", "check what's rotting in here". Scoped to a whole repository: not for ad-hoc bug hunts on a single file (use feature-dev:code-reviewer), and not for scoring an individual skill's quality (use skill-benchmark).

0
installs
▌
Model Advisor · chienchuanw
Use when you're about to start a task and want to pick the right Claude model and effort/thinking level for it, instead of defaulting one model + one effort to everything. Fires on "which model should I use", "what model and effort for this", "model advice", "model-advisor", "用哪個模型", "這個任務該用什麼模型/effort". Manual-only (`disable-model-invocation`). This is the CANONICAL model/effort rubric for the repo — `handoff-prompt` reuses the same table for its suggested-model line. Not a handoff tool (that is `handoff-prompt`); this just recommends a model + effort for the task in front of you.

0
installs
▌
Branch Report · chienchuanw bundle
Generates a branch-vs-default comparison report with plain-language explanations and senior dev review. Use when the user asks to summarize, compare, review, or explain branch changes, or wants a PR summary before merging. Read-only: it produces a report to read — it does not push the branch or open/update the PR (for that use gh-pr).

0
installs
▌
Pre Push Test · chienchuanw bundle
Use when the user wants to prevent broken code from being pushed, block git push when tests fail, run tests before push, set up push protection, or add a safety net against pushing untested code. Covers any request to gate pushes on test results — whether they mention "hook", "pre-push", or just describe wanting to stop bad code from reaching the remote. Also handles auto-fix/auto-repair of test failures before push. Trigger for: "block push if tests fail", "run tests before pushing", "protect branch from broken pushes", "stop pushing untested code", "catch failures before they reach remote", or any test-gated push workflow regardless of language (npm, pnpm, cargo, go, pytest, etc).

0
installs
▌
Spend Audit · chienchuanw
Intercept a hesitant purchase and run a structured Q&A audit to separate impulse / novelty / self-reward from a genuine recurring need — so the user avoids unnecessary spending. Reads the user's money-principles, asks a fixed backbone of behavioural questions (real frequency, real pain vs want, existing substitute, source of the urge, timing/opportunity-cost, card & bottom-line checks) via AskUserQuestion, reaches a verdict (buy / don't / use what you have / find cheaper / cool-down N days), and appends the decision to a running audit log. Use whenever the user is tempted by or unsure about a purchase and wants to check it's not impulse — "should I buy X", "想買 X 但猶豫", "該不該買", "help me audit this purchase", "幫我審這筆消費", "衝動消費", "spend audit", "consumption audit", "avoid an impulse buy" — even if they don't name a skill. This is the PRE-purchase sibling of daily-reviewer: daily-reviewer logs money already spent at day's end; spend-audit intercepts a purchase BEFORE it happens. For nightly 記帳 of spends that alread

0
installs
▌
Gmail Helper · chienchuanw
Act as the user's personal email secretary across all linked Gmail accounts (work, personal, semi, lighting, foufa) via the gmail-mcp server. Triage and label today's incoming mail, auto-applying existing labels on confident matches and proposing new labels with AskUserQuestion when nothing fits; then chase the status of older threads (awaiting reply, needs action, snoozed) and write the whole picture into the user's Obsidian daily journal. Use this skill whenever the user asks to "check my email", "do my inbox", "triage my mail", "what needs my attention", "go through my Gmail", "be my secretary", "any important emails", "follow up on old threads", or otherwise wants their multi-account inbox organized, labeled, or summarized into their journal — even if they don't say the word "skill" or name a specific account.

0
installs
▌
Auto Integrate · chienchuanw bundle
Use when you want an agent to autonomously take a finished feature branch through review, self-fix, and merge into an integration branch (`dev`) without a human reading the PR — the agent runs an objective code review, fixes or refactors what it can, gates on a hard green light (tests pass AND no blocking findings), then does a clean rebase-merge and logs a one-line revert entry. Fires on "auto-integrate this branch", "review-fix-and-merge to dev", "run the autonomous merge loop", "自主整合這個分支", "跑自動 merge 迴圈". Manual-only (`disable-model-invocation`) because it merges code. It NEVER touches `main`/production (release stays manual). Not a PR-review reporter (that is `/code-review`) and not the full issue→PR delivery flow (that is `feature`).

0
installs
▌
Handoff Prompt · chienchuanw
Companion to the `handoff` skill: after `/handoff` has written a handoff document, run this in the SAME (ending) session to emit a copy-paste English prompt the user pastes into a FRESH session so a new agent resumes the work seamlessly. Use when a session is wrapping up and the next session needs a clean bootstrap — ESPECIALLY right after a spec or plan is finalized/locked, when the context window is filling up or approaching its limit, or just before a `/clear` to continue the work elsewhere. Triggers: "give me the handoff prompt", "what do I paste into the new session", "產生接手 prompt", "handoff-prompt", "下一輪要貼什麼", "context 快滿了", "spec 定稿了幫我接棒", "要 /clear 了". Manual-only (`disable-model-invocation`). It does NOT write the handoff doc (that is `/handoff`) and does NOT launch a background agent (that is `claude-handoff`).

0
installs
▌
UI Visual Gate · chienchuanw
Use when starting or integrating work on a UI/visual project — anything that renders a screen, component, page, or layout a human looks at. Enforces a visual-first gate: for a UI project you confirm the look (a signed-off Figma frame or a throwaway preview page) BEFORE writing implementation code, and any UI change must carry a visual comparison (screenshot / preview URL / Figma frame) before it can merge. Fires when a project is flagged `ui_project: true` or the diff touches front-end files (.tsx/.jsx/.vue/.svelte/.css), and on "is the design signed off", "visual gate", "UI 專案要先確認畫面", "這個要附截圖對照嗎". Figma is the source of truth — this skill reuses the existing figma skills for design↔code sync rather than reimplementing them. The `auto-integrate` skill calls this as an extra green-light condition for UI projects.

0
installs
▌
Daily Planner · chienchuanw
Plan the user's day end-to-end. Triage all five Gmail accounts (by running gmail-helper), read Google Calendar and the Objectives MOC, carry forward yesterday's unfinished tasks, classify every task on an Eisenhower matrix (urgent × important), then write a prioritized plan with time-block suggestions into the Obsidian Daily note plus a reasoning log in the Journal. Use this whenever the user wants to plan, organize, or prioritize their day — "plan my day", "what should I do today", "make my daily plan", "set up my day", "morning routine", "help me prioritize my tasks", "規劃今天", "今天要做什麼", "排今天的任務", "幫我安排今天" — even if they don't name a skill. This is the morning-routine orchestrator: it runs gmail-helper as its own first step, so prefer it over gmail-helper whenever the user wants a whole-day plan and not only email triage.

0
installs
▌
Nutritionist · chienchuanw bundle
Act as the user's personal dietitian over their Obsidian vault. Reviews meals (from a photo or a text description), gives diet advice grounded in his health data (blood lipids, fatty liver, HbA1c, ECG T-wave, Ibuprofen allergy), body composition and target weight/body-fat, and continuously tracks intake toward those targets. Coach-type: it sets a daily calorie target + protein floor + the three cuts, scores each meal and each week against them, and self-calibrates the target when weight stalls. Diet-first — the lever a flat weight is missing. Dual-mode single skill: log/review a meal (instant, per-meal red/green verdict + one swap + running daily budget) OR advise / weekly review (ties the week's eating to weight-週均 + waist + the blood-lipid recheck and updates the O4 objective). Damage-control mode for meals he can't control (eating out, tour catering, family dinners): best-of-the-available + rebalance the week, never a demoralizing failure X. Use whenever the user sends a food photo or wants a meal reviewed

0
installs
▌
Daily Reviewer · chienchuanw
Close out the user's day with an honest retrospective. Read today's Daily note plan, interview the user to catch work done-but-unticked, then score completion, assess every active objective's on-track status, update the tracking files (tick checkboxes, carry unfinished tasks into 延後 with reasons, refresh an objective's 目前進度 when a milestone moved), and write a review into both the Daily note and the Journal with concrete suggestions for tomorrow. Use this whenever the user wants to look back on, review, wrap up, score, or close out their day, or asks whether they are on track — "review my day", "how did today go", "what did I get done", "daily retro", "wrap up my day", "am I on track", "end of day review", "回顧今天", "回顧一下今天", "今天做得如何", "今天完成了什麼", "睡前回顧", "檢討今天", "今天有沒有在軌道上" — even if they don't name a skill. This is the evening bookend to daily-planner: daily-planner builds the morning plan, daily-reviewer reviews it at night. Prefer this skill over daily-planner whenever the user is looking BACKWARD at a day t

0
installs
▌
Watch Video · chienchuanw bundle
Digests YouTube videos the user has queued but has no time to watch. For each video URL, pull the transcript (via yt-dlp — captions only, no video download) and dispatch a sub-agent that summarizes it, decides WHETHER IT IS EVEN WORTH WATCHING (skip / read-summary-only / skim-with-timestamps / watch-fully), fact-checks its load-bearing claims, draws conclusions, and — when it has learning value — maps it to the user's Objectives and projects. Then file an enriched reference note into Personal/Reference/, log one line in today's Daily note, and report a TL;DR led by the watch-or-skip verdict. The point is to STOP wasting time: most queued videos should come back "skip — here are the 3 lines." Use when the user pastes one or a few YouTube URLs and wants them digested, triaged, summarized, or filed — "should I watch this", "digest this video", "這支影片值得看嗎", "幫我看這支 YouTube", "把這支影片摘要歸檔", "這個影片重點是什麼" — even if they don't name a skill. Sibling of read-article (that skill is for web articles; this one is for YouTube v

0
installs
▌
Apply Coding Rules · chienchuanw
Stamp a managed coding-discipline rules block into a project's CLAUDE.md (or AGENTS.md / GEMINI.md) — the four rules: think before coding, simplicity first, surgical changes, goal-driven execution, plus a guard on irreversible/git-history actions. This skill is the single source of truth for those rules; running it writes or updates the rules between managed markers, idempotently, leaving the rest of the file untouched. Use when the user wants to add, update, sync, or propagate the shared coding rules / coding standards into a repository — "apply the coding rules", "add the four rules to this project's CLAUDE.md", "sync our coding standards here", "stamp the coding discipline block" — even if they don't name the skill.

0
installs
▌
Read Article · chienchuanw bundle
Reads web articles the user has no time to read. For each URL the user pastes, dispatch a sub-agent that fetches the article, summarizes it, fact-checks its core claims against the web, draws conclusions, and — when the article has learning value — works out how to apply it to the user's actual Objectives and projects. Then file an enriched reference note into the user's Obsidian vault (Personal/Reference/) and report a TL;DR back, plus a one-line learning record in today's Daily note. Use whenever the user pastes one or more article URLs and wants them digested, summarized, verified, classified, or filed — "read this for me", "summarize this article", "整理這篇文章", "幫我看這篇", "這篇幫我摘要歸檔", "把這幾篇讀一讀存進 vault" — even if they don't name a skill or say the word "skill".

0
installs
▌
Learning Coach · chienchuanw
Deliberate-practice / ultralearning coach over the user's Obsidian vault. Two modes. PLAN: given a goal or skill, decompose it into sub-skills, pick the highest-leverage weak point, define a feedback mechanism and 1-3 checkpoints, and write a learning plan into Personal/Learning/<slug>.md. CHECKPOINT: quiz the user with active-recall questions grounded ONLY in material THEY specify (a topic, a Reference note, a textbook chapter), grade their answers against that material, surface the gaps, set the next difficulty notch (+1), and append a checkpoint row to the learning file. The point is to verify the user ACTUALLY absorbed something — not that they "studied". Use when the user wants to plan how to learn something, set up deliberate practice, or be tested on whether they really understood material — "規劃學習 X", "幫我設計學習計畫", "怎麼學 X 比較有效", "考我", "查核我學會了嗎", "測我懂不懂", "active recall", "刻意練習查核", "設一個學習查核點", "quiz me", "test my understanding", "set a learning checkpoint". Enforces the TOEFL-8/12 gate and refuses to beco

0
installs
▌
Portfolio Review · chienchuanw
Read-only thinking check for the Obsidian portfolio at Personal/Finance/Portfolio/. Reads every open ticker file, compares the original opening thesis against later thesis-log entries and the most recent price action, and surfaces drift (you bought for reason A but are now justifying with reason B), anchoring (you keep restating the same thesis without re-examining it), and stale theses (no new entry in months despite price/news moves). Produces a markdown review the user can paste into their Journal — it challenges your *reasoning* and never tells you to buy or sell. Use this whenever the user wants to sanity-test their investing *thinking* rather than get moves — "is my thesis still right", "challenge my thinking on X", "am I anchoring", "are my reasons still valid", "thesis check", "portfolio retrospective", "重新檢視持股的理由", "我的持股邏輯還成立嗎". For actionable buy/sell/rebalance suggestions and directional calls — "what should I do", "any moves to make", "should I trim X", "投資建議" — use portfolio-advisor instead, whic

0
installs
▌
Portfolio Update · chienchuanw
Ingest broker / wallet screenshots and update the Obsidian portfolio tracker at Personal/Finance/Portfolio/. Reads each screenshot the user pastes (US broker, TW broker, crypto wallet/exchange), diffs the extracted holdings against existing ticker files, surfaces inferred buys/sells/transfers for user confirmation, prompts a one-line thesis-log entry for each change, then writes the transactions row, updates frontmatter (current_shares, avg_cost, last_price, last_price_date), and refreshes the portfolio.md MOC (fx_rates, allocation table, last_snapshot). Use this whenever the user wants to update their portfolio, log new trades, refresh prices, ingest broker screenshots, do a portfolio snapshot, or any phrasing like "update my portfolio", "I bought some X", "here's my latest holdings", "log this trade", "refresh portfolio prices", "更新投資組合", "對帳", "對一下股票". Pair skill of portfolio-review (evening / weekly check) — this skill WRITES state, portfolio-review READS it.

0
installs
▌
Portfolio Advisor · chienchuanw
Actionable, current-fact-grounded advice for the Obsidian portfolio at Personal/Finance/Portfolio/ — both structural calls (concentration vs. your targets, FX tilt, correlated bets) and directional calls on specific tickers (trim / add / hold), each grounded in a live price + news lookup and tied back to your written thesis. Chains portfolio-review as its thinking-check phase, then escalates to "what should I actually do next?" and writes a dated, gradeable advice memo. Use whenever the user wants moves, decisions, or suggestions rather than a pure reasoning check — "what should I do with my portfolio", "any moves to make", "should I rebalance", "what should I buy or sell", "give me portfolio suggestions", "投資建議", "現在該買賣什麼", "下一步該怎麼做", "幫我看看要不要調整". For a read-only check of whether your *thinking* is sound (drift / anchoring / stale thesis) with no buy/sell calls, use portfolio-review instead — this skill is the action-oriented sibling that makes the calls portfolio-review refuses to.

0
installs
▌
Gotcha Capture · chienchuanw
Use when the user encountered a problem, bug, edge case, or frustration with an existing skill and wants to save that knowledge into the skill's file. Trigger phrases include "add a gotcha", "document this pitfall", "capture lessons learned", "record edge cases", "pull out what went wrong", but also trigger when the user simply describes something that went wrong or was painful with a named skill (like readme, commit-msg, seo-meta, branch-report, gh-pr) and asks to update it, save it, or make sure it doesn't happen again. This skill mines conversation history for failure knowledge and writes it into the skill's Gotchas section. Do NOT use for creating new skills, adding features to skills, or changing how skills behave — only for documenting what went wrong.

0
installs
▌
Skill Benchmark · chienchuanw bundle
Benchmark a skill's quality by analyzing its structure, description, documentation, and live value-add from multiple perspectives using parallel agents. Use when the user asks to benchmark, evaluate, score, audit, assess, or rate a skill's quality, or wants to know if a skill is worth keeping vs. deprecating. Also trigger when the user asks "how good is this skill", "is this skill useful", "should I keep this skill", "review skill quality", or wants a thorough multi-dimensional analysis of a skill before deciding whether to improve or remove it. Scoped to one skill: for a whole-repository health/security audit use health-audit, not this. Takes a skill name as argument (e.g., /skill-benchmark readme). Produces a scored report with actionable improvement suggestions. Do NOT use for creating new skills, editing skill code, or capturing gotchas — those are different workflows.

0
installs

38 published skills