story-deck
Use when user needs to plan, storyboard, review, critique, or rewrite a presentation outline or slide deck, including SCR presentations, BBP/Beyond Bullet…
Skill by yysun
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/yysun-story-deck
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents, presentations
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/story-deck
What's inside
- Skill name
story-deck- Description
- Use when user needs to plan, storyboard, review, critique, or rewrite a presentation outline or slide deck, including SCR presentations, BBP/Beyond Bullet Points presentations, scene-based decks, headline development, bitmap visual generation, and handoffs to Markdown, Marp, or PPTX production.
- License
- MIT
- Instructions
- 207 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (10)
- SKILL.md
- openai.yaml · agents/
- bbp.md · references/
- handoff.md · references/
- headlines.md · references/
- scr.md · references/
- story-scenes.md · references/
- visuals.md · references/
- gemini_asset_lib.mjs · scripts/
- generate_gemini_asset_cli.mjs · scripts/
SKILL.md (first lines)
# Story Deck Strategist
Build persuasive presentation storyboards using one selected framework. Do not teach SCR or BBP back to the user; use the framework as an internal validation and output contract. Focus on the user's storyline, titles, visuals, notes, and artifacts.
## Core Contract
**Framework selection**
- Default to **SCR** when the user does not name a framework.
- Use **BBP** only when the user asks for BBP, Beyond Bullet Points, Atkinson-style, visual story, or persuasive visual storyboard.
- Do not mix SCR and BBP unless the user explicitly asks to combine them.
**Mandatory behavior**
- Select the central Story Thesis before creating slides.
- Build one deck-wide arc. Each slide is one scene beat, not a mini deck.
- In BBP mode, validate the natural business argument before assigning BBP labels.
- Treat examples by framework: SCR uses examples as support; BBP uses examples as scenes.
- Show a slide-title approval list before detailed storyboard work.
- Run the Title Refinement Pass before showing that approval list.
- Keep evidence precise; never invent metrics, quotes, customer facts, or financial figures.
- Label support claims in notes: `Fact:` · `Assumption:` · `Inference:` · `Evidence gap:`
**Artifact behavior**
- Auto-save Markdown for full storyboards, visual-first rewrites, critiques with reworked storyboards, Marp drafts, PPTX handoffs, and any explicit save/export request.
- If the user asks for visuals/images/assets, use the bundled visual CLI. Do not search for another image script and do not create one.
## Reference Loading
Read only what the request needs:
- Slide-title approval, quick outline, full storyboard, critique, rewrite, Marp, or PPTX: `references/headlines.md`
- Full storyboard, critique, rewrite, Marp, or PPTX: `references/story-scenes.md`
- SCR mode: `references/scr.md` for validation rules and allowed labels
- BBP mode: `references/bbp.md` for validation rules and allowed labels
- Bitmap visuals or generated image assets: `references/visuals.md`
- Markdown save, Marp, PPTX, or finished deck handoff: `references/handoff.md`
## Output Depth
- **Quick outline:** slide-title list only: `{#}. [{label}] {Title}`
- **Full storyboard:** Step 4 title approval, then Step 5 detailed storyboard
- **Critique existing deck:** Step 2 thesis, Step 3 ideal storyline, then Step 6 critique
- **Visual-first rewrite:** full storyboard with visual recommendations and bitmap briefs where applicable
## Workflow
### Step 0 - Context Discovery
Scan the working directory only when the user references the current project, provides local files, or asks to use repository context. Otherwise prioritize the supplied topic, pasted material, or external brief.
Use discovery to infer topic/product, audience, objective, source evidence, likely storyline, and artifact target. Do not ask for information already supplied or discoverable.
### Step 1 - Intake
If audience, goal, and output mode are clear, state assumptions and proceed. If key details are unclear, ask up to 3 questions:
| # | Question | Options |
|---|----------|---------|
| 1 | Who is this for? | Executives / Board · Investors · Sales / Prospects · Internal team |
| 2 | What should it achieve? | Get approval · Persuade / drive adoption · Inform or explain · Support evaluation |
| 3 | What output do you need? | Quick outline · Full storyboard · Critique existing deck · Visual-first rewrite |
If the topic, organization, product, or decision is still unclear, ask for that specifically.
### Step 2 - Story Thesis Discovery
Before creating slides, select the central thesis:
```text
Story Thesis: [One memorable business claim that anchors the deck.]
Audience Shift: [What the audience believes before vs. after.]
Core Tension: [The gap or conflict that creates momentum.]
False Belief: [What the audience currently assumes that the deck must correct.]
Turning Point: [The moment the audience sees the problem differently.]
Deck Metaphor: [Optional: one consistent metaphor/world for generated visuals.]
```
For strategic, executive, investor, product, recommendation, or BBP decks, generate 2-3 thesis candidates before selecting the strongest unless the user already supplied a strong thesis. For each candidate, include:
- Story Thesis
- Audience Shift
- Core Tension
- False Belief
- Turning Point
- Why it may hit
- Weakness / risk
Select the strongest thesis based on clarity, tension, business consequence, and memorability. If the user asked for an interactive checkpoint, ask them to choose. If the user asked for a finished artifact, state the selected thesis as an assumption and continue to the slide-title approval gate.
### Step 3 - Framework Storyline
Use the selected framework reference as the validation contract:
- SCR: validate the storyline against `references/scr.md` for SCQA, Pyramid, dot-dash, example-as-support behavior, and slide labels.
- BBP: first run the BBP Flow Validation Pass in `references/bbp.md`, then map the natural business flow into Act I/II/III, anchors/explanation/backup, example-as-scene behavior, and slide labels.
In BBP mode, output `Natural Business Flow -> BBP Label Mapping` before the formal BBP storyline. Output the framework storyline only once, then continue to Step 4. For critique-only requests, go to Step 6 after establishing the ideal storyline.
### Step 4 - Slide Title Confirmation Gate
Before working on slide details, propose the slide sequence as titles only and ask for approval. Do not continue to the detailed storyboard until the user approves the title list.
Before showing the title list, run the Title Refinement Pass in `references/headlines.md`. Present the refined titles, not the weaker first draft.
```text
Proposed Slide Titles:
1. [Framework label] [Plain-language title / headline]
2. [Framework label] [Plain-language title / headline]
3. [Framework label] [Plain-language title / headline]
...
```
Rules:
- Titles must reflect the Story Thesis, framework storyline, and deck-wide arc.
- In BBP mode, labels are metadata. Do not let `Hook`, `Relevance`, `Challenge`, or `Anchor` force a weaker sequence than the business argument wants.
- Use domain nouns, concrete contrast, and business consequences.
- For SCR, titles are Dots/action-title claims, not topics.
- For BBP, titles are reveal headlines that make the next scene necessary.
- In BBP Act II, anchors are arguments, not product history.
- SCR examples prove or clarify a Dot; BBP examples create concrete scene beats.From yysun/awesome-agent-world, MIT.
More from yysun/awesome-agent-world
- 0.519score
git-wiki
yysun
Build and maintain local code-project wikis under .wiki or language-specific wiki roots such as .wiki-cn and .wiki-fr.
SkillCFreshness: C★ 3Claude Code +9 - 0.510score
ai-workspace
yysun
Create, review, audit, and validate AI workspaces for agent hosts such as Codex, Copilot, Gemini, and similar desktop or CLI runtimes.
SkillCFreshness: C★ 3Claude Code +9 - 0.449score
notebooklm
yysun
Use this skill for NotebookLM tasks to create notebooks, add sources, query content, and generate artifacts through the NotebookLM CLI.
SkillDFreshness: D★ 3Claude Code +9 - 0.449score
react-app
yysun
Use when creating, modifying, refactoring, reviewing, or organizing a React web app that uses Tailwind or utility-first CSS.
SkillDFreshness: D★ 3Claude Code +9 - 0.443score
music-to-svg
yysun
Use the converter script to render a MusicXML file to SVG markdown image output.
SkillDFreshness: D★ 3Claude Code +9 - 0.440score
workspace-design
yysun
Design or review business operation pages, workbenches, and review or follow-up pages as task-centric workspaces.
SkillDFreshness: D★ 3Claude Code +9 - 0.424score
youtube-search
yysun
Use this skill when the user asks to search YouTube or find videos by keyword.
SkillDFreshness: D★ 3Claude Code +9
Signals
- Stars
- 3
- Forks
- 0
- Open issues
- 1
- Last commit (this folder)
- May 14, 2026 (5 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/yysun/awesome-agent-world
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from yysun
- 0.519score
git-wiki
yysun
Build and maintain local code-project wikis under .wiki or language-specific wiki roots such as .wiki-cn and .wiki-fr.
SkillCFreshness: C★ 3Claude Code +9 - 0.510score
ai-workspace
yysun
Create, review, audit, and validate AI workspaces for agent hosts such as Codex, Copilot, Gemini, and similar desktop or CLI runtimes.
SkillCFreshness: C★ 3Claude Code +9 - 0.449score
notebooklm
yysun
Use this skill for NotebookLM tasks to create notebooks, add sources, query content, and generate artifacts through the NotebookLM CLI.
SkillDFreshness: D★ 3Claude Code +9 - 0.449score
react-app
yysun
Use when creating, modifying, refactoring, reviewing, or organizing a React web app that uses Tailwind or utility-first CSS.
SkillDFreshness: D★ 3Claude Code +9 - 0.443score
music-to-svg
yysun
Use the converter script to render a MusicXML file to SVG markdown image output.
SkillDFreshness: D★ 3Claude Code +9 - 0.440score
workspace-design
yysun
Design or review business operation pages, workbenches, and review or follow-up pages as task-centric workspaces.
SkillDFreshness: D★ 3Claude Code +9
Cite this page
story-deck skill. Agent Mart. https://agentmart.dev/skills/yysun-story-deck (last updated 2026-10-24)