菜单
Generate professional slide deck images from content with smart logo branding. Use for creating slides, presentations, decks, or PPT output.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Extract visual, video, and audio features from media. Use for pixel features (Pillow), video frames (FFmpeg+Pillow), speech/audio features (openSMILE), music features (librosa), and visual semantic analysis (Gemini API batch or standard).
Orchestrate end-to-end academic manuscript preparation following APA 7th edition. Use for writing papers, drafting sections, or academic writing support.
Capture session decisions, conventions, and lessons into plan files, auto-memory, and CLAUDE.md so a fresh session resumes cleanly. Use for "hand off," "wrap up," "update docs for next session," or before /compact.
Run exploratory data analysis on tabular datasets with measurement-appropriate statistics. Use for EDA, descriptive statistics, data exploration, or preparing data summaries for reports and manuscripts.
Convert documents into business stories and infographics. Use for turning reports, documents, or text into visual summaries or infographics.
Stage and commit uncommitted changes with conventional commit messages. Use for committing changes or grouping commits by topic.
| name | q-presentations |
| description | Generate professional slide deck images from content with smart logo branding. Use for creating slides, presentations, decks, or PPT output. |
Generate slide deck images from content with layout-driven overlay safety and smart logo branding. Built on baoyu-slide-deck.
If in plan mode: write a brief plan — "Run q-presentations skill: analyze content, confirm style/audience preferences, generate outline and prompts, generate slide images, overlay logo, merge to PPTX/PDF." — then exit plan mode immediately. Do NOT attempt content analysis, image generation, or file merging while plan mode is active.
Agent execution instructions:
SKILL_DIR.${SKILL_DIR}/scripts/<script-name>.| Script | Purpose |
|---|---|
scripts/gen_slide.py | Generate slide images via GPT Image 2 (default) or Gemini |
scripts/overlay_logo.py | Apply Dr. Q logo overlay |
scripts/merge-to-pptx.ts | Merge slides into PowerPoint (Bun/TS) |
scripts/merge-to-pdf.ts | Merge slides into PDF (Bun/TS) |
Install: pip install openai google-genai Pillow python-dotenv | Bun runtime (via npx -y bun, Node.js 18+).
Default image model is gpt-image-2 (requires OPENAI_API_KEY). Set IMAGE_MODEL=gemini or pass --model gemini to gen_slide.py to use gemini-3-pro-image-preview (requires GEMINI_API_KEY).
16 presets available (default: blueprint). Full specs in references/dimensions/presets.md and references/styles/*.md.
Full step details: references/workflow_detailed.md
| Step | Action | Key references |
|---|---|---|
| 0 | Skill announcement | — |
| 1 | Analyze content, load EXTEND.md preferences | references/analysis-framework.md, references/config/preferences-schema.md |
| 2 | Confirm style, audience, video overlay, logo | references/options.md |
| 3 | Generate outline with layout selection | references/outline-template.md, references/layouts.md |
| 4 | Review outline (conditional) | — |
| 5 | Generate per-slide prompts | references/base-prompt.md, references/content-rules.md |
| 6 | Review prompts (conditional) | — |
| 7 | Generate slide images | scripts/gen_slide.py |
| 7.5 | Logo overlay | scripts/overlay_logo.py |
| 8 | Merge to PPTX/PDF | scripts/merge-to-pptx.ts, scripts/merge-to-pdf.ts |
| 9 | Output summary | — |
For editing existing decks: references/modification-guide.md. For audience and visual hierarchy: references/design-guidelines.md.
Output directory: slide-deck/{topic-slug}/ containing source-{slug}.{ext}, outline.md, prompts/ (numbered .md files), numbered slide .png files, {topic-slug}.pptx, and {topic-slug}.pdf.
Detection priority: --lang flag > EXTEND.md > user conversation language > source content language. All user-facing responses use the user's preferred language; technical names remain in English.
Include: Slide deck generation from content with logo branding and video-overlay-aware layouts. Exclude: Slide editing in PowerPoint, animation, speaker notes.