Review Image

Review screenshots or other images with OpenRouter vision models via bundled Deno scripts. Use for quick VRT sanity checks, invalid-image screening, or CI gates. `scripts/review-image.ts` returns freeform feedback; `scripts/review-image-ci.ts` returns strict `pass|fail` JSON and exits non-zero on fail.

423 stars 6 forksImages, Video & Audio

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add mizchi/skills --skill review-image
Manual install
git clone --depth 1 https://github.com/mizchi/skills.git /tmp/skills
mkdir -p ~/.claude/skills
cp -r /tmp/skills/review-image ~/.claude/skills/review-image

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# review-image

Use this skill when you want a cheap vision-model opinion on an image before doing heavier work.

## When to use

- Quick VRT sanity checks
- Detect obviously broken, blank, malformed, or suspicious inputs
- Ask a model whether a screenshot or generated image matches a prompt
- Add a simple image-review gate to CI

## Requirements

- Deno
- `OPENROUTER_API_KEY` or `OPENROUTER_KEY`

## Invocation

Both scripts are executable Deno files with shebang `#!/usr/bin/env -S deno run --allow-env --allow-net --allow-read`. Run them directly — no `deno run` wrapper needed.

When the skill is APM-installed, the scripts live under `~/.claude/skills/review-image/scripts/`. Examples below use the `~/...` form, which is shell-expanded. In non-shell exec contexts (e.g. `child_process.spawn` without `shell: true`) both `~` and `$HOME` stay literal — substitute the resolved absolute path (`/Users/<you>/.claude/skills/review-image/scripts/...`).

Flags can appear before, between, or after the positional args. `--model` accepts both `--model X` and `--model=X`. Both scripts write program output to **stdout** and logs/errors to **stderr** — but only the JSON-emitting modes (CI gate always; freeform with `--json`) are safe to pipe straight into `jq` / `JSON.parse`. Freeform default mode is prose, not JSON.

## Scripts

### Freeform review — `review-image.ts`
Read the full skill on GitHub

Details

Skill name
review-image
Repository
mizchi/skills
Path
review-image/SKILL.md
Repo updated
18 November 2026

FAQ

What does the Review Image skill do?

Review screenshots or other images with OpenRouter vision models via bundled Deno scripts. Use for quick VRT sanity checks, invalid-image screening, or CI gates. `scripts/review-image.ts` returns freeform feedback; `scripts/review-image-ci.ts` returns strict `pass|fail` JSON and exits non-zero on fail.

How do I install the Review Image skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add mizchi/skills --skill review-image. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Review Image popular?

It lives in mizchi/skills, which has 423 GitHub stars and 6 forks.

Related skills

Actions CI Tuning

423

mizchi/skills

Use when auditing or improving GitHub Actions workflows for a project. Covers cache setup (npm/pnpm/yarn), job parallelism, shard-based test splitting, artifact handling pitfalls, and Playwright-specific patterns. Trigger on: slow CI, cache miss, flaky shard jobs, merge-reports failures, or an explicit 'tune CI' request.

Testing & QA

AI Index

423

mizchi/skills

Method and tooling for measuring how AI-generated a piece of prose reads, in Japanese or English. Two layers — a deterministic lint for removable surface tells (scripts/slopscore.py, no API) and a judgment layer of typed probabilistic questions (Jev noul/score) for the tells that survive editing. Includes measured human/AI baselines, a gap-first calibration procedure, and labeled fixtures. Use when building or calibrating an AI-likeness score, or when a draft needs a defensible number rather than a vibe. For writing guidance use natural-writing-ja / natural-writing-en instead.

AI, LLMs & Agents

Apm Usage

423

mizchi/skills

Reference for APM manifests, installation, targeted updates, deployment paths and lockfiles. Invoke when the user explicitly asks about APM or apm.yml, or an unfamiliar APM command/error needs exact CLI details. Do not automatically invoke for every project that happens to have installed skills.

DevOps & Cloud

Boring Video

192

sugarforever/boring-video-studio

Use when developing a content-driven video from a topic, research bundle, article, or draft, or when routing an existing script or non-argument video into HyperFrames production.

Images, Video & Audio28

Performance Expert

423

mizchi/skills

Performance specialist perspective for the weekly review. Focuses on bundle size, LCP / CLS / INP, avoidable re-work, image and font optimization. Reads audit-bundle and audit-lighthouse raw output when available.

Images, Video & Audio

Record Terminal

192

sugarforever/boring-video-studio

Use when a video must show terminal commands, CLI output, streamed execution, full-screen terminal applications, or code entered inside a terminal session.

Images, Video & Audio28

Browse the most popular Claude skills, all images, video & audio skills, or write your own with the skills generator.