What is the difference between Interactive and Auto mode?

Interactive mode presents findings one by one for user approval before any action is taken. Auto mode asks clarifying questions upfront and then proceeds to fix or skip issues autonomously using best-effort decision criteria.

Can I filter findings by severity?

Yes, you can pass arguments like 'high', 'medium', or 'all' to the command to filter which findings are processed based on their priority level.

How does the skill handle Test-Driven Development (TDD)?

When the project has a detected test framework, the skill follows a Red-Green-Refactor workflow: writing a failing test, implementing the fix, and then verifying the results before proceeding.

What happens if a session is interrupted?

The skill saves its progress to a .handle-cr-state.json file. If the session is interrupted or the context is compacted, you can resume exactly where you left off.

Handle Code Review Findings

Name: Handle Code Review Findings
Author: alphaleonis

byalphaleonis

0•

Collaboration Tools

Resolves code review findings interactively or autonomously using Test-Driven Development workflows and intelligent state tracking.

The handle-cr skill streamlines the post-review development cycle by parsing structured code review documents and guiding the AI through each identified issue. It offers a flexible workflow with a default interactive mode for manual oversight and an autonomous mode for high-velocity fixing. By leveraging existing test infrastructure, it prioritizes a Red-Green-Refactor TDD approach to ensure every fix is verified. With built-in state persistence, users can recover long-running sessions after context compaction, making it an essential tool for maintaining code quality in complex projects.

Key Features

01Pattern-based grouping to fix similar architectural issues in batches

02Persistent state tracking via JSON for session recovery

03Automated Test-Driven Development (TDD) workflow integration

04Granular severity filtering for Critical, High, and Medium findings

05Interactive and Autonomous (Auto) resolution modes

060 GitHub stars

Use Cases

01Automating the TDD cycle for behavioral bug fixes identified during peer reviews

02Systematically resolving security and performance findings after a codebase audit

03Applying consistent refactoring patterns across multiple files simultaneously

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add alphaleonis/decaf-claude-config handle-cr

For use in Claude.ai and ChatGPT

Key Features

01Pattern-based grouping to fix similar architectural issues in batches

02Persistent state tracking via JSON for session recovery

03Automated Test-Driven Development (TDD) workflow integration

04Granular severity filtering for Critical, High, and Medium findings

05Interactive and Autonomous (Auto) resolution modes

060 GitHub stars

Use Cases

01Automating the TDD cycle for behavioral bug fixes identified during peer reviews

02Systematically resolving security and performance findings after a codebase audit

03Applying consistent refactoring patterns across multiple files simultaneously

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add alphaleonis/decaf-claude-config handle-cr

For use in Claude.ai and ChatGPT