Skip to main content
Review

Get a code review that ranks issues by severity

Ask Gemini to review a diff for correctness, security, and clarity, rank findings by severity with a concrete failure scenario each, and skip style nits unless they hide a bug.

提示词

Review this change for correctness, security, and clarity. For each finding: give a severity (blocker / major / minor), the specific failure scenario, and the smallest fix. Skip pure style nits unless they hide a bug. If the change looks safe, say so plainly. Diff: [DIFF].

替换变量

[DIFF]

The diff or the changed files.

该用哪个模型

  • Gemini 3.1 ProSeverity judgement and failure scenarios need careful reasoning.

示例输入

  • [DIFF] = a pull request touching auth middleware.

示例输出

我们只发布来自真实模型运行的示例输出。此模板尚未经过正式测试,因此不显示输出。请用上方的输入自行运行。

为何这一结构有效

  • Severity + failure scenario filters out noise and false alarms.
  • Permission to say "looks safe" reduces invented problems.

常见失败与修复

Every line gets a comment.

修复:Cap it: "Report at most the 5 most important findings."

已知局限

  • A model review complements, not replaces, human review and tests.

与 Google 无隶属关系,也未获其背书。Gemini、Nano Banana 和 Veo 是 Google LLC 的商标。