code-review:review-local-changes

Review local uncommitted git changes for bugs, security, and maintainability.

Updated Mar 28, 2026
One-click install
npx skills add https://github.com/DITEKmax/rutcampustrack --skill code-review-review-local-changes-ditekmax
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review:review-local-changes
Source: https://github.com/DITEKmax/rutcampustrack/tree/main/.claude/skills/review-local-changes
Command: npx skills add https://github.com/DITEKmax/rutcampustrack --skill code-review-review-local-changes-ditekmax

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured, pre-commit review of local uncommitted changes to catch bugs, security vulnerabilities, and contract or quality regressions before code enters version history.

Core Features & Use Cases

  • Staged vs unstaged awareness: Differentiates and reports whether issues come from staged or unstaged diffs so developers know what is ready to commit.
  • Parallel specialist agents: Launches multiple focused agents (security, bug hunting, contracts, test coverage, historical context, code-quality) to surface diverse issue categories.
  • Impact and confidence filtering: Scores each finding by impact and confidence and applies configurable thresholds (e.g., --min-impact) to reduce noise and focus on actionable issues.
  • Machine-friendly output: Optionally emits a structured JSON report for programmatic CI consumption or a human-readable markdown summary for local review.

Quick Start

Run a local review focusing on security and performance with medium impact threshold and JSON output.

Frequently Asked Questions about code-review:review-local-changes

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review local uncommitted changes before a git commit?

Review local uncommitted changes by running an automated pre-commit code review that analyzes staged and unstaged diffs to identify bugs, security vulnerabilities, and maintainability issues before code enters version history.

Can I output code review results as JSON for CI pipeline consumption?

Yes, you can output code review results as JSON for CI pipeline consumption. The review process optionally emits a structured JSON report for programmatic CI usage or a human-readable markdown summary for local developer review.

How do I filter automated code review findings to focus on high-impact issues?

Filter automated code review findings by applying configurable impact and confidence thresholds, such as the min-impact flag, to reduce noise and focus only on actionable issues scored by multiple parallel specialist agents.

Does the pre-commit code review differentiate between staged and unstaged diffs?

Yes, the pre-commit code review differentiates between staged and unstaged diffs. It explicitly reports whether identified issues originate from staged or unstaged changes so developers know exactly what is ready to commit.

What is the best way to run a security audit on local code changes?

The best way to run a security audit on local code changes is to launch parallel specialist agents focused on security, bug hunting, and code quality to surface diverse issue categories and compute confidence scores for each finding.