ワンクリックで
build-slidev-decks
Fast validation of Slidev deck changes. Correct invocation, output interpretation, and common issues.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Fast validation of Slidev deck changes. Correct invocation, output interpretation, and common issues.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Review an exec-talk's README and produce an exec.recipe.yml. Uses a 3-agent collaborative council to analyze section weighting, executive audience fit, narrative arc, and action clarity. Always overwrites any existing recipe. Triggers: "review the talk", "create the recipe", "is this landing for execs", "section weighting", "coverage gap", "recipe", "executive framing".
Review a tech-talk's README and produce a deck.recipe.yml. Uses a 3-agent collaborative council to analyze section weighting, narrative arc, and coverage gaps. Always overwrites any existing recipe. Triggers: "review the talk", "create the recipe", "is this the best use of our time", "section weighting", "coverage gap", "recipe".
Read from and write to the Workbench project memory store. Use when starting complex content work (query for prior context) or finishing a session that changed content (record drawer entries). Never invoke during agent pre-flight gates.
Add persona-grounded audience framing to a tech-talk. Selects 2–3 relevant personas, adds a "Who This Hits" box and up to 3 reaction lines to the README, and adds persona speaker notes to the slide deck. Triggers: "add personas", "who this hits", "persona reactions", "audience framing".
Use this skill when asked to verify, validate, fact-check, or review any data entry, lore fact, character bio, episode record, quote, or show content for accuracy against canonical project sources and the live database. Triggers on: check data, verify data, fact-check, accuracy check, data accuracy, check entry, verify entry, review accuracy, is this accurate, canon check, lore check, check this data, verify this, review this entry, check for errors, validate data, is this right, is this canon.
Use this skill when asked to verify, validate, or fact-check a lore entry, character description, seed data record, or any Breaking Bad content against the canonical reference in docs/breaking-bad-universe.md. Triggers on: check lore, verify lore, fact-check, canon check, lore accuracy, is this correct, seed data accuracy.
| name | build-slidev-decks |
| description | Fast validation of Slidev deck changes. Correct invocation, output interpretation, and common issues. |
| infer | true |
Quick validation protocol for Slidev presentation changes.
Load this skill when you need to:
.md file changes produce valid buildsslides/Windows PowerShell:
cd C:\Users\hobob\source\CopilotTraining\slides
Linux/macOS Bash:
cd /path/to/CopilotTraining/slides
chmod +x build.sh scripts/build-all.sh # First time only
| Goal | Windows | Linux/macOS |
|---|---|---|
| Validate single deck | .\build.ps1 -Deck copilot-cli | ./build.sh --deck copilot-cli |
| Validate tech-talks | .\build.ps1 -Folder Tech-talks | ./build.sh --folder tech-talks |
| Validate workshop | .\build.ps1 -Folder workshop | ./build.sh --folder workshop |
| Validate exec-talks | .\build.ps1 -Folder exec-talks | ./build.sh --folder exec-talks |
| Rebuild all | .\build.ps1 | ./build.sh |
[DONE] line[DONE] 17 presentations built, 0 archived skipped.
✅ If you see this, builds passed. Done. ❌ If count is lower, one or more decks failed. See troubleshooting.
When editing a single .md file or component that affects only one deck, build just that deck to save time:
# Windows: build only copilot-cli
.\build.ps1 -Deck copilot-cli
# Also verbose for debugging:
.\build.ps1 -Deck copilot-cli -Verbose
# Linux/macOS: build only copilot-cli
./build.sh --deck copilot-cli
The script will:
.md file[DONE] and [CLOCK] linesWhen to use:
.md file or its frontmatter[SCIENCE] Building tech-talks slides...
[HAMMER] tech-talks/agent-teams... [OK] 4.3s
[HAMMER] tech-talks/agentic-journey... [OK] 2.3s
...
[HAMMER] tech-talks/vscode-latest... [OK] 4.2s
[DONE] 17 presentations built, 0 archived skipped.
[CLOCK] Total time: 1m 0.8s
node.exe : [Slidev] Ignored provided index.html warnings in stderr[OK] status on the next line[FAILED] or count is wrongCheck the failing deck:
# Windows
.\build.ps1 Tech-talks -Verbose
# Linux/macOS
./build.sh tech-talks --verbose
Look for error message in output. Usually:
<template> tagFix and retry — the build command.
Permission denied on Linux/macOSchmod +x build.sh scripts/build-all.sh
./build.sh tech-talks
First build after major dependency changes. Clear cache and retry:
npm cache clean --force
npm install
# Then run build command
Verify your builds using these expected counts:
| Category | Expected Count | Approx Time |
|---|---|---|
| Tech-talks | 17 | ~60s |
| Workshop | 7 | ~30s |
| Exec-talks | 4 | ~20s |
| All decks | 28 | ~2m |
If you need to script or automate builds:
| Aspect | Windows | Linux/macOS |
|---|---|---|
| Category case | Tech-talks | tech-talks |
| Verbose flag | -Verbose | --verbose |
| Script location | slides/build.ps1 | slides/build.sh |
Both wrapper scripts (build.ps1, build.sh) automatically handle path resolution. Never run from scripts/ directly — always invoke from slides/ directory. The wrappers detect their own location and navigate for you.