# Edit product review and unboxing videos with Framefields

A useful product review answers a buying question with visible evidence. An unboxing can introduce the item, but the test is what makes the video useful: show what happened, under what conditions and who the product might suit.

Framefields can organize your footage into a repeatable review format. Keep the actual observations in a data file and let the composition handle chapter cards, comparison labels and the final verdict.

## Start with one test

For a 45-second desk microphone review, record the contents of the box, the setup and two audio samples. Use the same distance and room for both samples. Document the microphone settings before editing.

| Time | Picture | Evidence |
| --- | --- | --- |
| 0–4 s | Microphone in use | Preview the buying question |
| 4–10 s | Box contents | Show what is included |
| 10–18 s | Connection and setup | Reveal the practical friction |
| 18–35 s | Two labeled samples | Let the viewer compare |
| 35–45 s | Verdict card | Best use and one limitation |

A portrait version works for a short review. If the comparison needs longer listening, make a landscape version with full samples instead of cutting away the evidence.

## Keep sound comparisons honest

Use `Layer.video` for footage and `Layer.text` for test labels. During an audio comparison, remove the music bed and avoid processing one sample differently from the other. If you normalize samples, tell the viewer and apply the same stated method.

Store the product name, test conditions and observations separately from the layout. Use [charts](/docs/guides/charts-and-media/#charts) only for measurements you actually made. A subjective impression should stay a labeled observation, not become a fabricated numerical score.

For ordinary unboxing footage, cuts and restrained text animation are enough. A different effect on every shot can make the viewer work harder to inspect the product.

## Use Codex, ChatGPT, Claude, Cursor or Grok Bot

You can use **Codex, ChatGPT, Claude, Cursor, or Grok Bot** to plan the edit and write Framefields TypeScript. In a coding environment with project and terminal access, ask the assistant to implement the brief and run the checks. In a chat-only environment, ask for the source files, save them locally and run them in your Framefields project. Rendering needs Node.js 22+ and a supported WebGPU environment; see [getting started](/docs/getting-started/).

For Codex or Cursor, install the [Framefields skills](/docs/agents/install/):

```bash
npx skills add gatewai-dev/framefields -a codex -a cursor
```

For Claude Code, use `/plugin install framefields`. With ChatGPT or Grok Bot, supply the [agent reference](https://framefields.dev/docs/llms.txt), your brief and asset list; available execution tools depend on your environment. The workflow does not require a native integration in every assistant.

Create a project with `npx framefields init my-video --install --pm npm`, then work inside `my-video`. Use the generated font setup and put your source media under `assets/`. Run `npm run typecheck`, inspect `comp.lint()` findings, render selected frames and a grid, then run `npm run preview` before `npm run render`. The [agent workflow](/docs/agents/workflow/) covers inspection and QA. Check portrait layouts against the destination app's current controls before uploading.

## Give your assistant the evidence

```text
Build a 45-second Framefields microphone review, 1080x1920, 30 fps.
Use assets/shots.json and assets/review-notes.md. Include unboxing,
setup, two complete labeled audio samples, then a verdict with one
strength and one limitation. Music must stop during the samples.
Keep both samples at the supplied levels; do not add enhancement.
Never invent measurements, prices or a recommendation absent from
the notes. Include the supplied sponsorship disclosure if applicable.
Show a preview with sound before exporting.
```

## Review the verdict against the footage

Play the samples on headphones and a phone speaker. Make sure the labels stay on the right sample after every cut. Confirm that the verdict follows the test rather than overstating it.

When you review the next item, reuse the scene structure and swap the evidence. If your template assumes every product has three advantages, change the template; a review should fit what you found.

For a brand-owned promotional version, use the separate [UGC-style ad workflow](/blog/ugc-style-video-ads/).