What happens if a fix cannot be implemented automatically?

The skill tracks the status of every item and provides a summary report at the end, highlighting failed items with specific error details and suggested manual follow-up actions.

Can I see the changes before they are applied?

Yes, you can use the --dry-run argument to generate a prioritized list of findings and planned actions without modifying any files.

Does it work with external libraries or just Claude plugins?

It handles both. It uses specialized logic for Claude Code components and leverages MCP research for external frameworks, APIs, and general application code.

How does the skill know what to fix?

Apply Findings scans your current conversation context for actionable items like audit logs, review comments, or numbered lists of improvements generated in previous turns.

Apply Findings

Name: Apply Findings
Author: melodic-software

bymelodic-software

•

Developer Tools

Automates the implementation of recommendations and fixes identified during code audits or reviews within your conversation.

Apply Findings is a powerful productivity skill for Claude Code that bridges the gap between identifying issues and fixing them. It intelligently scans your current conversation for actionable items—such as audit findings, code review suggestions, and prioritized TODOs—and generates an execution plan categorized by severity. The skill employs a smart research phase that automatically consults relevant documentation or MCP tools to ensure fixes follow best practices, then systematically applies changes with built-in verification and reporting, making it an essential tool for rapid, high-quality code remediation.

Key Features

0138 GitHub stars

02Intelligent research layer for both Claude-specific and general code patterns

03Detailed summary reporting of implemented, skipped, and failed items

04Automated verification of code modifications and configuration syntax

05Dry-run capability to preview changes before implementation

06Prioritized remediation logic (Critical through Info severity levels)

Use Cases

01Automatically remediating security vulnerabilities found during an /audit-style command

02Applying comprehensive architectural improvements suggested during a detailed code review

03Updating Claude-specific skills and agents based on the latest ecosystem documentation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add melodic-software/claude-code-plugins apply-findings

For use in Claude.ai and ChatGPT

Download Skill