coco-fix-iterate

Automates iterative code refinement cycles combining ML and AI technologies with a focus on quality, safety, and efficiency.

1|1|Updated Feb 1, 2026
One-click install
npx skills add https://github.com/corbat-tech/coco --skill coco-fix-iterate
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: coco-fix-iterate
Source: https://github.com/corbat-tech/coco/tree/main/.claude/skills/coco-fix-iterate
Command: npx skills add https://github.com/corbat-tech/coco --skill coco-fix-iterate

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates the process of improving code quality by iteratively reviewing, scoring, fixing, and verifying code until a target quality threshold is met, reducing manual effort and ensuring higher standards.

Core Features & Use Cases

  • Iterative Quality Improvement: Continuously refines code based on objective scoring across multiple dimensions (correctness, security, coverage, etc.).
  • Multi-Agent Architecture: Employs specialized agents for review, fixing, and verification to prevent bias and enhance objectivity.
  • Configurable Quality Targets: Allows users to set specific score thresholds and iteration limits.
  • Use Case: Automatically refactor a legacy codebase to meet modern security and maintainability standards, or ensure a new feature adheres to a strict 95% test coverage requirement before merging.

Quick Start

Run the coco-fix-iterate skill to improve the quality of the entire project, aiming for a score of at least 90.

Frequently Asked Questions about coco-fix-iterate

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

FAQPage Schema
How do I automate continuous code quality improvement until a target score is met?

You can automate iterative code quality improvement by running a convergence loop that continuously reviews, scores, fixes, and verifies code until meeting configurable target metrics like correctness and security. This reduces manual effort and ensures higher standards.

What is a multi-agent workflow for automated refactoring and code review?

A multi-agent workflow for automated refactoring uses specialized agents to independently handle code review, fixing, and verification cycles. This architecture prevents bias and enhances objectivity when evaluating code across multiple dimensions like security and test coverage.

Can I configure score thresholds and iteration limits for automated code review?

Yes, you can configure specific score thresholds and maximum iteration limits for the code review process. This allows you to target precise quality metrics, such as a strict 95% test coverage requirement, before stopping the convergence loop.

Does the iterative code fixing loop support a single-agent mode for rapid improvements?

Yes, the iterative code fixing loop supports a single-agent mode alongside its multi-agent architecture. This mode is designed for focused or rapid improvements when you need faster convergence without running the full multi-agent review cycle.

When should I use an autonomous code quality convergence loop instead of manual refactoring?

Use an autonomous code quality convergence loop when refactoring a legacy codebase to meet modern security standards or ensuring new features adhere to strict coverage requirements. It automates repetitive review and fixing cycles, significantly reducing manual refactoring effort.