skills.rest
swat9013
swat9013COMMUNITY
Explore all skills by this author →
GitHub

review-fix

Orchestrates automated review and correction cycles for source code.

Updated Aug 27, 2026
Search Tags:#code review#subagents#orchestration#quality gate#batching#automated fix
One-click install
npx skills add https://github.com/swat9013/dotfiles --skill review-fix-swat9013
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: review-fix
Source: https://github.com/swat9013/dotfiles/tree/main/.claude-global/skills/review-fix
Command: npx skills add https://github.com/swat9013/dotfiles --skill review-fix-swat9013

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates the repetitive and error-prone cycle of post-implementation code review and correction by orchestrating isolated review and fix subagents, reducing manual overhead and improving code quality and consistency.

Core Features & Use Cases

  • Orchestration of isolated reviews: Delegates context-isolated reviews to an Opus subagent across four dimensions (Architecture, Test Strategy, API Design, Behavior).
  • Automated fixes and gating: Delegates code modifications to a Sonnet subagent, runs lint/test quality gates, and enforces batching rules for large change sets.
  • Reporting and traceability: Produces a structured review-fix.md with per-issue IDs, recurring issue detection, and cycle history for PR or repository-level quality workflows.
  • Use Case: Run this on a pull request to automatically detect changed files, iterate up to three review/fix cycles, apply non-recurring automated fixes, and produce a final audit report.

Quick Start

Run the review-fix skill on the target repository path to detect changed files, run isolated reviews, apply automated fixes where safe, and emit a review-fix.md summary.

Frequently Asked Questions about review-fix

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

FAQPage Schema
How do I automate code review and correction cycles for pull requests?▼

Automated code review and correction cycles can be orchestrated by delegating context-isolated reviews to an Opus subagent across architecture, tests, API design, and behavior, while a Sonnet subagent applies fixes and runs quality gates.

What is the best way to run isolated code reviews on changed files in a repository?▼

Isolated code reviews on changed files use a subagent to evaluate source code across four dimensions: architecture, test strategy, API design, and behavior, ensuring context isolation for thorough analysis without interference.

How does automated code review handle large change sets and batching?▼

Automated code review handles large change sets by enforcing batching rules that group changed files, allowing a Sonnet subagent to apply non-recurring automated fixes safely while running lint and test quality gates.

Can I generate a structured audit report for automated code fixes applied to a PR?▼

Yes, a structured review-fix.md report can be generated featuring per-issue IDs, recurring issue detection, and cycle history to provide traceability for automated code fixes applied during pull request quality workflows.

What are the limitations of automated review-and-fix cycles for source code?▼

Automated review-and-fix cycles are limited to a maximum of three iterations and only apply non-recurring automated fixes, meaning persistent issues may require manual intervention if they reappear across multiple cycles.

Do I need subagents to orchestrate post-implementation code quality gating?▼

Yes, orchestrating post-implementation code quality gating requires context-isolated subagents to separately handle review detection across multiple dimensions and apply automated fixes before running lint and test validation.

Related Skills

Explore More Software Engineering Skills→
B0yZ4kr14B0yZ4kr14

speckit-security-review-branch

Analyze branch or PR diffs for security vulnerabilities across injection, secrets, and authorization domains.

Community
Intermediate
robertzengcnrobertzengcn

gstack-openclaw-investigate

Investigate application bugs through a five-phase root cause analysis workflow.

Community
Advanced
CarbeneAICarbeneAI
9

subagent-driven-development

Dispatches fresh subagents per task with spec and quality reviews.

Community
Advanced
shaharKeisarAppsshaharKeisarApps

technical-researcher

Analyzes code quality, architecture, and documentation gaps in target repositories.

Community
Basic