# Review

> Run a code review using Cursor Agent with Gemini

- Skill: `kieldro/review` (Agent Skill)
- Install (CLI): `npx skillmds@latest add kieldro/review`
- Raw SKILL.md: https://api.skillmd.com/api/skills/kieldro/review/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: Kieldro (https://skillmd.com/u/kieldro)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/kieldro/review

---


# Code Review with Cursor Agent

Run a comprehensive code review using Gemini 3.1 Pro via cursor-agent.

## Instructions

1. Determine the target: if `$ARGUMENTS` is provided, review those specific files. Otherwise, review all uncommitted changes (staged + unstaged).
2. Build the file list and read the relevant code.
3. Run cursor-agent with Gemini to review:

```bash
cursor-agent --model gemini-3.1-pro --mode ask --print --trust \
  "Review the following code for bugs, race conditions, error handling, security, and UX issues. Rate each area 1-10 and give an overall score. Be concise. Files: <file list>" \
  --workspace "$(pwd)"
```

4. Present the results to the user with a summary table.

## Usage

- `/review` — review all uncommitted changes
- `/review transcribe-daemon.py` — review a specific file
- `/review src/` — review all files in a directory