gbro-collage-broll
Skill that turns a short voiceover line into halftone paper-collage B-roll through Gemini
Medium risk
We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.
Why this level
- Runs local Python scripts and calls the Gemini API
- Video generation is billed and spends money on the API
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/gbro-collage-brollDetects the agents on your machine, checks the risk and pins the version.
Other ways to install
Run in a terminal in the project folder
npx skills add pyang5166/gbro-collage-broll --skill gbro-collage-broll -a claude-code -yThe skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.
Run in a terminal in the project folder
npx skills add pyang5166/gbro-collage-broll --skill gbro-collage-broll -a cursor -yThe skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.
Run in a terminal in the project folder
npx skills add pyang5166/gbro-collage-broll --skill gbro-collage-broll -a github-copilot -yThe skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.
Run in a terminal in the project folder
npx skills add pyang5166/gbro-collage-broll --skill gbro-collage-broll -a codex -yThe skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.
Run in a terminal in the project folder
npx skills add pyang5166/gbro-collage-broll --skill gbro-collage-broll -a gemini-cli -yThe skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.
Run in a terminal in the project folder
npx skills add pyang5166/gbro-collage-broll --skill gbro-collage-broll -a cline -yThe skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.
Run in a terminal in the project folder
npx skills add pyang5166/gbro-collage-broll --skill gbro-collage-broll -a roo -yThe skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.
A fork of Roo Code, same .roo folders.
Run in a terminal in the project folder
npx skills add pyang5166/gbro-collage-broll --skill gbro-collage-broll -a opencode -yThe skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.
Install the skill: git clone https://github.com/pyang5166/gbro-collage-broll.git ~/.agents/skills/gbro-collage-broll. Then run scripts/check_setup.sh and configure GEMINI_API_KEY, ffmpeg and Python as prompted.
Other ways from the author
git clone https://github.com/pyang5166/gbro-collage-broll.git ~/.agents/skills/gbro-collage-brollFor Claude Code clone into ~/.claude/skills/gbro-collage-broll. After install run scripts/check_setup.sh.
This is third-party code. Review the repository files before installing.
What it does
The skill takes a voiceover line of about five seconds and builds a short clip in an editorial halftone paper-collage style. It runs through three required approval gates: first it proposes only the visual metaphor, then after your sign-off it renders a color still frame, and only then it generates the video. The final animation is made by the Gemini Omni Flash model from a first and last frame, with frame-by-frame quality checks. By default it delivers a vertical 720x1280 clip, five seconds, silent, ready to sit under narration.
Who it is for. For video creators, editors and SMM specialists who need background footage under a voiceover.
Good fit when
- You need a short B-roll clip under a specific voiceover line
- You want one collage style across a series of clips
- You want to approve the frame idea before spending on generation
Not a fit when
- No access to the Gemini API or you do not want paid generation
- You need audio or a clip longer than a few seconds
- You need a realistic look rather than a paper collage
Example request
collage b-roll: people think AI thinks for you, but it is more of a mirror that shows the holes in your questionLimitations
You need a GEMINI_API_KEY from Google AI Studio, and video generation is billed per use. Gate 2 with the still frame relies on the built-in image_gen tool of the Codex environment. It also needs Python 3.10 or newer, google-genai, and ffmpeg with ffprobe. The default model is fixed to gemini-omni-flash-preview.
How to disable. Remove the gbro-collage-broll folder from your agent skills directory.
Security check
- Runs local Python scripts and calls the Gemini API
- Video generation is billed and spends money on the API
README in short
The README shows finished samples and describes a three-gate protocol: metaphor, still frame, video. The generation and upload scripts ship with the skill, so nothing extra needs installing. Requirements are listed: a Codex environment, a Gemini key, Python, google-genai and ffmpeg. Installation is just cloning the folder into the agent skills directory. MIT licensed.
SKILL.md
--- name: gbro-collage-broll description: 将约 5 秒口播文稿、观点句或抽象概念做成高级 editorial halftone paper-collage / 半调纸拼贴 B-roll。用户说“collage b-roll”“纸拼贴 b-roll”“半调拼贴”“拼贴风格配画面”“用这段文稿做拼贴动画”“gbro-collage-broll”，或希望把一句文稿转成拼贴视觉隐喻时，必须使用此 skill。强制采用三阶段审批：先只提视觉隐喻，用户确认后才生成彩色拼贴静帧，静帧再次确认后才默认调用 Gemini Omni Flash 生成首尾帧组装动画。默认视频模型固定为 gemini-omni-flash-preview，不再默认使用 Veo；只有用户明确指定其他模型时才切换。 compatibility: 在 Codex 环境运行（Gate 2 依赖内置 image_gen）。视频生成脚本已随本 skill 自带（scripts/generate_video.py），另需 Python >= 3.10、google-genai >= 2.10.0、已配置的 GEMINI_API_KEY，以及 ffmpeg / ffprobe。首次使用先按「首次使用：环境自检」完成配置。 --- # gbro Collage B-roll 把一句约 5 秒的口播压成一个 sharp visual idea，再做成高级编辑风纸拼贴组装动画。 默认链路： 1. 只设计视觉隐喻，等待用户确认 2. 只生成最终静帧，等待用户确认 3. 自动调用 Gemini Omni Flash 生成视频并完成 QA
FAQ
Why two manual approvals?
A wrong metaphor or a bad still frame going straight to video wastes real API money. Editing text in Gate 1 is free, and re-rendering a frame in Gate 2 is cheaper than re-running a clip.
Can I switch the video model?
It defaults to gemini-omni-flash-preview. Another model is used only if you name it explicitly.
Related
Skills and CLI from HeyGen: the agent writes HTML compositions and renders them into MP4 videos, captions, motion graphics and decks
Blender MCP
MCP for Blender
An MCP server and Blender addon: the agent creates and edits objects, materials and scenes, pulls assets and runs Python in Blender
Official Remotion skills for making videos in React: compositions, animation, captions, maps and rendering
Agentic video studio for Claude Code, Cursor and Codex: script, voiceover, images, stock footage, music, subtitles and render