Generate a 10-clip b-roll pack for a talking-head video
One agent prompt that reads your transcript and writes ten generation prompts for cutaways that show what is being said, each tagged with the sentence it covers.
Read the transcript below. Find the 10 sentences where a cutaway would help most (a concrete object, place or action is mentioned, and nobody's lips need to be seen). For each, write a video generation prompt for a 5-second b-roll clip that literally shows what the sentence says: subject, action, setting, shot size and one camera move, light, the style lock below, ambient sound, 'No on-screen text, no people's faces.' Output a table: sentence (quoted), timestamp, prompt, aspect ratio {9:16}. STYLE LOCK: {style lock paragraph} Transcript: {paste transcript with timestamps}
In Claude Code, paste the prompt as a message, or run /video-edit which does this end to end.
Steps
- Get the transcript with timestamps first (the transcribe skill or any STT tool).
- Generate the ten clips; name files by the timestamp (0m42s.mp4) so the editor finds the slot.
- Place each cutaway on its own sentence, never over a sentence where the speaker's lips are visible on camera.
Tips
- Faceless b-roll avoids the character consistency problem entirely; keep people's faces out of generated cutaways.
- Measured on a real edit: 12 of 54 automatically chosen cutaways were off topic. Read every prompt against its sentence before generating.
| works with | Claude Code: Runs as a skill or a prompt in the terminal; can read your files and write the result. · Claude: Paste into claude.ai or the Claude app. · ChatGPT: Paste into ChatGPT; works with any GPT model. · Any LLM: Model-agnostic: Gemini, Mistral, a local model. |
| skill | npx skills add fortunto2/solo-factory/video-edit · source |
| cost | Free. No API, no key. |
| status | Unverified: written from the technique, not yet run end to end by us. |
| license | CC BY 4.0 for the text here; skills are MIT. |
| updated | 2026-11-01 |
| for agents | b-roll-pack-generation-prompt.md |
Related
Ask an agent to turn a script into a numbered generation shot list
A prompt for Claude or ChatGPT that converts a 30-second script into 6-8 shot prompts in the 7-part structure, ready to paste into a video model one by one.
B-roll shot list from a transcript: the table to film or generate from
An agent prompt that reads a transcript and returns a shot table (timestamp, line, shot, source: have / film / generate) for a camera or a video model, grouped so each location is filmed once.
Establishing shot: open a video with a place
A prompt for the wide opening shot that tells the viewer where they are, with the three things that make a generated landscape look filmed rather than rendered.
Cutaway matching: sentence to a literal visual, same place and time
A prompt that reads each sentence of a transcript and names the one picture that shows it, with the place and time it must come from, so the shot list is a list of things, not moods.
The day as one text file: an agent reads the speech and proposes the story
The step before any timeline: the day's speech in shooting order, numbered by clip and sentence, read by an agent like a scriptwriter. It proposes story, scenes, hook and cutaways; you argue in text.
Anime style: 2D cel look, limited animation, clean line weight
A filled 7-part prompt for a hand-drawn anime shot: flat cel shading, animation on twos, a held painted background, and the three things the style breaks in current video models.