Review Video

Video Review

Review videos by extracting frames and analyzing them visually.

Workflow

Step 1: Extract frames

python3 ~/.claude/skills/review-video/scripts/extract_frames.py  --smart -n 10 --cleanup

Step 2: Read ALL frames in parallel

Read every frame path from output using Read tool. Parallel reads for speed.

Step 3: Cleanup temp files

Run the cleanup command printed at the end (if --cleanup was used).

Step 4: Provide structured review

## Video Review: [filename]

**Metadata**: [resolution] | [duration] | [fps] fps | [codec]

### Overall Assessment
[1-2 sentences]

### Frame-by-Frame Notes
- > [0:00] Opening: [observation]
- o [0:05] [observation]
- * [0:10] Scene change: [observation]
- = [0:15] Closing: [observation]

### Issues Found
- [ ] Issue at [timestamp]: [description]

### Verdict
[PASS / NEEDS FIXES / FAIL] - [reason]

Options

| Flag | Description | |------|-------------| | -n NUM | Number of frames (default: 10) | | --smart | Use scene detection (recommended) | | --threshold FLOAT | Scene sensitivity 0.0-1.0 (default: 0.1, lower=more sensitive) | | --cleanup | Print cleanup command for temp files | | --json | Output as JSON | | -o DIR | Custom output directory | | -q NUM | JPEG quality 1-31 (default: 2, lower=better) |

Frame Source Icons

| Icon | Meaning | |------|---------| | > | Start of video | | = | End of video | | * | Scene change detected | | o | Evenly-spaced filler |

Frame Count Guidelines

| Use Case | Command | |----------|---------| | Quick check | -n 5 | | Standard | -n 10 --smart | | Animation | -n 20 --smart --threshold 0.05 | | Debug | -n 30 --smart |

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: floomhq
• Source: floomhq/moto
• License: MIT
• Homepage: https://github.com/floomhq/moto#readme

Install and usage instructions live in the source repository linked above.