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。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/alecs5am/ralphy --skill storyboard-generation命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
How to make an analog-horror PSA short — the stenciled-pictogram / robo-broadcast / VHS-overlay format (faceless, 10 scenes × ~3s, 9:16, ~30s total). A domain overlay on the standard pipeline that supplies the IF / DO-NOT / BUT / AND scenario structure, the locked yellow-on-black 1970s-civil-defense pictogram visual STYLE, the chroma-key-to-alpha rule (icons must be transparent PNGs before composition), the robo-PSA voice profile (ElevenLabs "Alerter" community voice + ALL CAPS input + stability ~0.5 + style 0), the layered VHS-noise overlay stack (SnowCanvas + VcrTrackingCanvas + MobiusScanlines + MobiusWobble), the 5-layer RGB chromatic-split for icons + captions, the SMPTE colour-bars climax (3 wide vertical bars + filter blur(2.5px) mandatory), and the `-tune grain` CRF 30 final-encode rule for noise-heavy renders. Works for ANY PSA topic — "your fridge is not your fridge", "your phone is not your phone". USE WHEN the user asks for an analog horror PSA / "creepy PSA-style short" / fake-emergency-broadcast
Prompt and asset generation craft — turns an approved scenario into prompts.json and generated media through `ralphy generate`. Owns anchor order (location-master-plate first), character fit, photoreal-human prompting, model choice and cost preview, the reference-photo policy, regeneration rules, and the image/video quality gate. USE WHEN the user asks to "generate prompts", "generate assets", "make images / video / voiceover / music", "regenerate scene-XX", "try a different model", "A/B variant", "how much will this cost". TRIGGER (EN): "generate prompts", "make the images", "regenerate scene", "model swap", "cost preview", "A/B variant", "redo this shot".
Audio-first long-form explainer pipeline — takes one audio file (or a YouTube/podcast URL → audio) and produces a faceless overlay-driven video in the dev-essay / tech-podcast style (5-30 minutes, 16:9, no talking head). Orchestrates the existing `ralphy` primitives: yt-dlp pull → silence-remove → ElevenLabs Scribe v1 word-level transcript → Gemini audio-describe → LLM claim segmentation → per-claim overlay-type planner (code-block / terminal / tweet-card / browser-frame / screenshot / meme / diagram / quote-card / chapter-card / logo-pop) → asset prep (Playwright for screenshots, `ralphy generate image` for memes, ElevenLabs Music for the bed, ElevenLabs SFX for whoosh/pop/hit) → HyperFrames composition with chapter sub-compositions → `ralphy render`. USE WHEN the user drops an audio file (mp3 / wav / m4a) or a YouTube / podcast URL and asks for a long-form video edited on top, says "make a video from this audio / podcast", "monter ce podcast" (FR), "smontiruy podkast v video" (RU transliteration), "audio to
基于 SOC 职业分类
正在显示 SKILL.md
| name | storyboard-generation |
| description | Rules and prompts for generating structured presentation slides and multi-panel storyboards. |
| license | MIT |
When generating full storyboard slides or presentation decks (typically using imagegen_2_0), follow these rules:
Strict Containment: The prompt must explicitly instruct the model to fit all photo cards, text blocks, and UI elements entirely within the slide boundaries. Nothing should be cropped or cut off at the edges. Use phrases like: "All photo cards fully inside slide boundaries, nothing cropped." Resolution & Aspect Ratio: Use aspect_ratio: "16:9" (or the requested slide format) and resolution: "1k" or "2k" to ensure text legibility and clean, sharp layouts. 1080p is an invalid resolution for imagegen_2_0 — never use it. Layout Definition: Clearly delineate rows and columns in the prompt structure (e.g., LAYOUT: 2 rows. Row 1: two equal columns (E1 left, E2 right)...) so the model organizes the elements logically. Reference Adherence: If replicating or modifying an existing layout, explicitly instruct the model to maintain the exact typography, icon set, card style, and background color. Timeline Bars: When visualizing a continuous shot or multi-scene sequence, explicitly prompt for a gradient timeline bar at the very bottom (e.g. Timeline bar at very bottom: gradient line from left (slow) to right (fast), labeled...). Consistency: Keep the lighting, time of day, and environmental context consistent across the panels on a single slide unless explicitly contrasting states (like a Before/After). Complex Storyboards (Film/Video): When building multi-panel film storyboards, use a structured JSON prompt with strict formatting directives (e.g., dark charcoal background #1a1a1a, white 2px borders around panels, specific typography for labels). Use cyan for camera movement overlays, yellow italic for audio, green for transitions. Embed a structured data block below each panel (Shot Type, Lens, Camera Path, Audio, Transition). Camera Movement Overlays: Explicitly prompt for arrows and diagrams drawn directly on the image panels (e.g., "curved cyan arc arrow showing orbit", "bold downward arrow labeled VERTICAL PLUNGE") to simulate professional director storyboards. Empty Panels for Re-use: When you need a storyboard panel that serves only as a structural placeholder (e.g., removing a character but keeping the text/layout), prompt the model for an explicitly empty dark frame: "COMPLETELY EMPTY dark panel — solid dark charcoal background, NO photo, NO person, NO image. Just the dark background with a subtle dark grey rectangle placeholder outline."