Lov Any2deck

PPT 大师 · PPT Master

Transform content into professional slide deck images.

Usage

/lov-any2deck path/to/content.md
/lov-any2deck path/to/content.md --style sketch-notes
/lov-any2deck path/to/content.md --audience executives
/lov-any2deck path/to/content.md --lang zh
/lov-any2deck path/to/content.md --slides 10
/lov-any2deck path/to/content.md --outline-only
/lov-any2deck  # Then paste content

Script Directory

Agent Execution Instructions:

1. Determine this SKILL.md file's directory path as SKILL_DIR
2. Script path = ${SKILL_DIR}/scripts/<script-name>

Options

Slide Count by Content Length:

Style System

Presets

Style Dimensions

Full specs: references/dimensions/*.md

Auto Style Selection

Design Philosophy

Decks designed for reading and sharing, not live presentation:

• Each slide self-explanatory without verbal commentary
• Logical flow when scrolling
• All necessary context within each slide
• Optimized for social media sharing

See references/design-guidelines.md for:

• Audience-specific principles
• Visual hierarchy
• Content density guidelines
• Color and typography selection
• Font recommendations

See references/layouts.md for layout options.

File Management

Output Directory

slide-deck/{topic-slug}/
├── source-{slug}.{ext}
├── outline.md
├── prompts/
│   └── 01-slide-cover.md, 02-slide-{slug}.md, ...
├── 01-slide-cover.png, 02-slide-{slug}.png, ...
├── {topic-slug}.pptx
└── {topic-slug}.pdf

Slug: Extract topic (2-4 words, kebab-case). Example: "Introduction to Machine Learning" → intro-machine-learning

Conflict Handling: See Step 1.3 for existing content detection and user options.

Language Handling

Detection Priority:

1. --lang flag (explicit)
2. EXTEND.md language setting
3. User's conversation language (input language)
4. Source content language

Rule: ALL responses use user's preferred language:

• Questions and confirmations
• Progress reports
• Error messages
• Completion summaries

Technical terms (style names, file paths, code) remain in English.

Workflow

For the full step-by-step workflow, use references/workflow.md. Keep the main flow in this order:

1. Setup & analyze: load .lov-skills/lov-any2deck/EXTEND.md when present, save source content, analyze style signals, detect language, choose slide count, and check for existing slide-deck/{topic-slug} output before continuing.
2. Confirm: use AskUserQuestion for style, audience, slide count, outline review, and prompt review. If custom dimensions are selected, collect texture, mood, typography, and density.
3. Generate outline: read the selected preset from references/styles/ or combine dimension docs from references/dimensions/, then write outline.md using references/outline-template.md.
4. Review outline when requested, then generate per-slide prompts under prompts/ using references/base-prompt.md and references/layouts.md.
5. Review prompts when requested, generate slide images, optionally apply branding with scripts/apply-branding.py, then merge PPTX/PDF with the TypeScript scripts.
6. Report the output directory, generated slide count, PPTX/PDF paths, style, audience, language, and any partial failures.

Partial workflows (--outline-only, --prompts-only, --images-only, --regenerate) and slide modification procedures are documented in references/workflow.md.

References

Notes

• Image generation: 10-30 seconds per slide
• Auto-retry once on generation failure
• Use stylized alternatives for sensitive public figures
• Maintain style consistency via session ID
• Step 2 confirmation required - do not skip (style, audience, slides, outline review, prompt review)
• Step 4 conditional - only if user requested outline review in Step 2
• Step 6 conditional - only if user requested prompt review in Step 2

Extension Support

Custom configurations via EXTEND.md. See Step 1.1 for paths and supported options.

Runtime context (shared)

运行前读取本 Skill 包的 skill.yaml，由宿主提供 skill-runtime/v1 上下文。字段解析顺序为：当前请求、项目上下文、个人 Preferences、品牌 Profile、通用默认值。

• 只使用 Manifest 声明的字段；Profile 保存公开品牌事实，Preferences 保存个人工作偏好。
• required: true 字段缺失时，按 Manifest 的问题配置向用户提出一个聚焦问题；用户明确同意后再保存回答。
• 报错提供可复制的 context_id、字段路径与来源，诊断内容避开秘密、完整私人路径和原始配置。

通用反馈闭环

用户在 Skill 驱动任务中提出修改意见时，继续当前产物前必须执行：

1. 先判断意见是 task-specific（仅本次）还是 reusable（可跨任务复用）。
2. task-specific 只修改当前任务，不改 Skill。
3. reusable 先确定作用域：领域规则先更新对应 canonical Skill；适用于所有 Skill 的规则先更新共享规范。
4. 完成规则更新、版本、lint 与分发核验后，再把修改应用到当前任务。
5. reusable 修改会使此前的“确认”“继续”“发吧”失效；完成当前产物修改和回读后必须停下，等待用户下一步指示，不自动进入发布、提交或其他外部写入。