good-broll

Generate B-roll image prompts from video scripts and voiceover copy.

45|4|Updated Jun 14, 2026
One-click install
npx skills add https://github.com/hongfamonvAI/good-broll --skill good-broll-hongfamonvai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: good-broll
Source: https://github.com/hongfamonvAI/good-broll
Command: npx skills add https://github.com/hongfamonvAI/good-broll --skill good-broll-hongfamonvai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires Pillow, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? Short-video creators struggle to turn abstract script ideas—concepts, comparisons, workflows, emotions—into concrete visuals. This Skill converts voiceover scripts, talking-head copy, and storyboard text into ready-to-paste image generation prompts for B-roll, covers, and explainer illustrations. ## Core Features & Use Cases - Script-to-Visual Analysis: Extracts core concepts, audience, actions, and information relationships from video copy, then maps abstract ideas (e.g., "efficiency gains") to concrete scenes (e.g., before/after comparisons). - Five Built-in Visual Styles: Editorial collage, warm hand-drawn illustration, playful 3D cartoon, whiteboard doodle, and warm sketch explainer (lo-fi SaaS style), each with detailed prompt templates in references/style_templates.md. - Copy-Ready Prompt Output: Produces complete prompts with composition, characters, props, UI elements, decorative motifs, and aspect ratio parameters for direct use in image generation models. - Use Case: A creator making an AI-tools tutorial pastes a script about building workflows 10x faster; the Skill analyzes the before/after relationship and outputs a collage-style prompt contrasting complex coding with drag-and-drop modules. ## Quick Start Paste your video script and ask the agent to generate B-roll illustration prompts for it, optionally naming one of the five visual styles.

Frequently Asked Questions about good-broll

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate B-roll image prompts from a video script?▼

Paste your video script or voiceover copy directly into the conversation. The Skill extracts core concepts, identifies visual relationships like before/after contrasts, selects a suitable style, and outputs a complete prompt you can copy into an image generation model.

What visual styles are available for short-video illustrations?▼

Five styles are built in: editorial collage (cut-paper social media look), warm hand-drawn illustration, playful 3D cartoon, whiteboard doodle, and warm sketch explainer for SaaS and workflow concepts. Detailed templates for each live in references/style_templates.md.

Can I use the generated prompts with Midjourney or other image models?▼

Yes. The output prompts are written as complete, copy-ready text including composition, characters, colors, and an aspect ratio parameter such as --ar 3:4, which is compatible with Midjourney-style parameter syntax and adaptable to other image generators.

Does this Skill generate images directly?▼

No. It produces text prompts for external image generation models, not the images themselves. The only local image generation is scripts/generate_showcase.py, which uses Pillow to render style showcase previews without any external image API.

When should I use the warm sketch explainer style instead of collage?▼

Use warm sketch explainer for product concepts, AI workflows, and SaaS feature explanations where minimal composition and black marker lines fit better. Collage style suits lively social-media tutorials, efficiency comparisons, and younger audiences.