video-comparer

An agent skill by daymade, from daymade/claude-code-skills. Tags: analysis, performance, qa, validation, video.

What it does

This skill should be used when comparing two videos to analyze compression results or quality differences. Generates interactive HTML reports with quality metrics (PSNR, SSIM) and frame-by-frame visual comparisons. Triggers when users mention "compare videos", "video quality", "compression analysis", "before/after compression", or request quality assessment of compressed videos.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add daymade/claude-code-skills --skill video-comparer

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/daymade/claude-code-skills
cp -r claude-code-skills/video-comparer ~/.claude/skills/video-comparer

Safety box score

Grade C, moderate risk: 28 of 100 risk points, rated 2026-09-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This skill runs FFmpeg/FFprobe on user-supplied videos, extracts frames, and writes an HTML comparison report plus frame image folders. The main risks are external command execution, system-wide FFmpeg installation instructions, writable output anywhere you point it, and a generated report that loads third-party assets from a CDN.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	2 (moderate)	It invokes external binaries with `subprocess.run(...)` for `ffmpeg` and `ffprobe`, which is core to the skill's purpose.
Network access	1 (low)	The report template loads JS/CSS from `https://unpkg.com/...`, so opening the report makes browser requests to a third party.
Filesystem reach	2 (moderate)	`scripts/compare.py` writes the HTML report and copies frames into `output_path.parent / 'original'` and `... / 'wechat'`, so it can create files and folders wherever the user points it.
Secrets & credentials	0 (none)	The files do not read, request, or transmit passwords, tokens, browser sessions, or `.env` contents.
Privilege & persistence	2 (moderate)	The docs instruct system-wide installs such as `brew install ffmpeg` and `sudo apt install ffmpeg`, which may require elevated privileges or modify the host package state.
Instruction hijack surface	0 (none)	I did not find instructions to ignore confirmations, follow fetched remote instructions, or hide behavior.
Irreversible actions	0 (none)	It generates a local comparison report only; it does not publish, post, deploy, pay, or delete anything on behalf of the user.
Transparency	0 (none)	The shipped code and template are readable; there are no binaries, base64 blobs, or runtime-downloaded installer scripts in the skill files.

Before installing

Check `assets/template.html` and decide whether loading JS/CSS from `unpkg.com` is acceptable for your environment.
If you do not want host-wide changes, install FFmpeg in a sandbox or container first instead of following the `sudo apt install ffmpeg` path.
Run the script once on sample videos and point `-o` to a scratch directory, since it creates `original/` and `wechat/` folders next to the report.

Rated 2026-09-20 by an automated review of 7 files from the skill's repository.

Source

Repository: daymade/claude-code-skills (all skills from this repository)
Path: video-comparer/SKILL.md
Branch: main
Updated: 2026-09-20

principle-fix-root-causes — Apply when debugging.
principle-prove-it-works — Apply after completing a task, before declaring done.
seo-page — Deep single-page SEO analysis covering on-page elements, content quality, technical meta tags, schema, images, and performance.
qa-expert — This skill should be used when establishing comprehensive QA testing processes for any software project.
web-perf — Audit, diagnose, or optimize website loading and interaction performance, Core Web Vitals, and Lighthouse performance scores.
aws-database — Routes any task involving AWS databases — choosing, comparing, recommending, getting started with.