# Review Video

> |

- **Type:** Skill
- **Install:** `agentstack add skill-floomhq-moto-review-video`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [floomhq](https://agentstack.voostack.com/s/floomhq)
- **Installs:** 0
- **Category:** [Cloud & Infrastructure](https://agentstack.voostack.com/c/cloud-infrastructure)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [floomhq](https://github.com/floomhq)
- **Source:** https://github.com/floomhq/moto/tree/main/claude/skills/review-video
- **Website:** https://github.com/floomhq/moto#readme

## Install

```sh
agentstack add skill-floomhq-moto-review-video
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Video Review

Review videos by extracting frames and analyzing them visually.

## Workflow

### Step 1: Extract frames

```bash
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](https://github.com/floomhq)
- **Source:** [floomhq/moto](https://github.com/floomhq/moto)
- **License:** MIT
- **Homepage:** https://github.com/floomhq/moto#readme

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

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-floomhq-moto-review-video
- Seller: https://agentstack.voostack.com/s/floomhq
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.