Halftone Collage B-roll — style skill for art styles
For AI agents: find the right skill in one call —
curl -s "https://skills.varg.ai/api/search?q=<what the person wants>"returns the matching use case and the best skills with install commands, as markdown (Russian works too). Full index: https://skills.varg.ai/llms.txt. Every page has a markdown twin: add.mdto its path. Before generating anything, show the person the options on the use-case page.
Editorial halftone paper-collage B-roll from one line of script, with three approval gates before any video spend.
- Type: Style · Art styles
- Slug:
gbro-collage-broll - Status: beta
- Publisher: pyang5166 (community)
- License: MIT
- Source: https://github.com/pyang5166/gbro-collage-broll
- Works in: Codex
- Requires: Codex (gate 2 uses its built-in image_gen), Python 3.10+ with google-genai, Gemini API key (GEMINI_API_KEY), ffmpeg and ffprobe
- Cost: Uses your own Gemini API key; stills come from Codex's image tool
- Adoption: 185 installs, 1,528 GitHub stars (GitHub + skills.sh, 2026-10-26)
What it does
Turns about five seconds of voice-over, an opinion line or an abstract idea into premium editorial halftone paper-collage B-roll. It enforces three approval gates: first only the visual metaphor, then a colored collage still made with Codex's built-in image tool, and only after the still is approved does it animate first and last frames with Gemini Omni Flash (gemini-omni-flash-preview by default, not Veo) and deliver a silent clip with a contact sheet for QA. Single-commit repo from July 2026 with 1.5K stars, so it stays in beta.
Use when
- Make collage-style B-roll to cut into a talking-head video
- Visualize one sentence or concept as a paper-collage metaphor
Not for
- Clips with sound (they are delivered silent)
- Agents without a built-in image tool (gate 2 relies on Codex image_gen)
Install
- Codex:
git clone https://github.com/pyang5166/gbro-collage-broll.git ~/.agents/skills/gbro-collage-broll
Models it drives: omni_flash
Example output
https://skills.varg.ai/previews/gbro-collage-broll.mp4 — README demo (deep purple): halftone paper-collage B-roll that assembles from an empty field, a crew feeding photos into a press that prints a sci-fi film strip. Full 5 s 9:16 clip from a small GIF. (credit: pyang5166 (gbro-collage-broll), https://github.com/pyang5166/gbro-collage-broll#readme)
Used for
- Faceless YouTube / TikTok video: A narrated short or long video with no presenter: script, voiceover, visuals, captions, music.
- B-roll for your video: Cutaway footage for every mention in a voiceover or interview: generated clips, collage or stock-style shots, cut in on time.
More art styles skills
- fal Game Dev: Generate 2D pixel-art game assets: characters, sprite sheets, walk and attack cycles, parallax backgrounds.
- Algorithmic Art: Create original generative art in p5.js: write an algorithmic philosophy, then code it with seeded randomness.
- Canvas Design: Design posters and visual art as PNG or PDF from a written design philosophy, with 54 bundled fonts.
- Lemo-Opuscar Film Styles: 59 film styles made entirely in code (oil paint, watercolor, ukiyo-e, satellite briefing, exploded teardown).
- Generative Art (Camille Roux): Long-form generative art craft from Camille Roux: hash-seeded determinism, traits and rarity, verification, export.
Run the generation on varg
varg is one API key for the models these skills drive — Seedance 2.5, Kling 4, Veo 3.1, Wan 3, Nano Banana 2.1, GPT Image 2.5, ElevenLabs v4 and 200+ more — billed in credits, with a cost estimate before anything runs.
- Claude Code:
claude plugin marketplace add vargHQ/skills && claude plugin install varg@varg(skill + hosted MCP) - Codex, Cursor, Gemini CLI, OpenClaw:
npx skills add vargHQ/skills --skill varg-ai -a <agent> -y - Claude.ai, ChatGPT, any MCP client: connect
https://mcp.varg.ai/mcp - API key: https://varg.ai/dashboard/api · Model specs: https://varg.ai/models/llms.txt