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.
For each sentence below, write the ONE cutaway that shows it literally. Output a table: # | Sentence | Visual (the object, place or action named in the sentence, as a shot: size + what is in frame) | Must come from (the place and the time window: same location, within {15} minutes of when the sentence was said) | On camera? (yes if the speaker's lips are visible in the source for this sentence; then the answer is 'no cutaway'). Rules: - Literal, not associated. 'The bakery was closed' is a shot of the closed bakery door, not a street. - If the sentence names nothing visible (an opinion, a feeling), write 'none' unless the previous or next sentence names something; then the cutaway spans both. - Never propose a shot from a place the story has not reached or has left. - One visual per sentence; if two things are named, pick the one the sentence is about. - Mark sentences where the thing named is shown nowhere in the footage list as MISSING, so it can be filmed or generated. Footage available (clip, place, time, one-line description): {paste shot list} Sentences (clip, time, text; 'lips visible' flag if known): {paste sentences}
In Claude Code, paste the prompt as a message, or run /video-edit which does this end to end.
Steps
- Produce the footage list from the analysis sidecars or by scrubbing: clip, place, shooting time, one line of what is visible. Without it the agent proposes shots you do not have.
- Run the prompt, then build the contact sheet of every chosen cutaway with its sentence under it and read it. Measured on one film: 12 of 54 automatic cutaways were off topic.
- For the MISSING rows decide: film it next time, or generate a faceless clip (see the generated-vs-real rule).
Tips
- The 'must come from' column is what separates a cutaway from stock footage. If you cannot fill it, the shot does not belong in a film about a day.
- Place cutaways in silent windows only. The 'on camera?' column exists so lips are never covered.
| 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 | cutaway-matching-sentence-to-visual-prompt.md |
Related
The cutaway rule: what is said, same place and time, never over moving lips
Three constraints for every cutaway in a talking film, and the agent prompt that applies them to a transcript. Kills what viewers read as broken: a picture from elsewhere, a face out of sync.
A contact sheet of every cutaway with its line under it
The one-page check that catches off-topic cutaways before anyone watches: every cutaway frame in a grid, its sentence under it. Measured once: 12 of 54 automatic cutaways were off topic.
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.
From a sentence to five search queries and one generation prompt
A prompt that turns each line needing a cutaway into five queries for your own footage or a stock site, ordered from literal to loose, plus one generation prompt as the fallback when nothing is found.
A mentioned object must be seen, ideally just before the line about it
The rule that turns 'X is mentioned, never shown' from a viewer complaint into a check you run on the transcript: every named thing gets a shot, placed just before or on the line.
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.