skills.rest
Tromml
TrommlOFFICIAL
Explore all skills by this author →
WebsiteGitHub

kimchi:review

Reviews implementation plans using multiple AI personas for actionable feedback.

8|Updated Jan 31, 2026
Search Tags:#planning#code review#complexity analysis#test coverage#scope management#persona review
One-click install
npx skills add https://github.com/Tromml/kimchi --skill kimchi-review
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: kimchi:review
Source: https://github.com/Tromml/kimchi/tree/main/plugins/kimchi/skills/review
Command: npx skills add https://github.com/Tromml/kimchi --skill kimchi-review

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the challenge of ensuring software development plans are robust, free from scope creep, and appropriately complex by leveraging multiple AI personas for a comprehensive review.

Core Features & Use Cases

  • Multi-Persona Critique: Employs five distinct AI personas (feature-trimmer, complexity-detector, premature-optimization-detector, scope-guardian, test-coverage-advocate) to scrutinize a plan.
  • Structured Feedback: Generates detailed concerns with clear recommendations (CUT, DEFER TO V2, SIMPLIFY, KEEP).
  • Interactive Refinement: Allows users to accept, reject, or modify persona recommendations, facilitating collaborative plan improvement.
  • Use Case: After generating an initial implementation plan for a new feature, use this Skill to have specialized AI agents identify potential issues like scope creep or unnecessary complexity before development begins.

Quick Start

Run the kimchi review skill on the current plan.

Frequently Asked Questions about kimchi:review

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

FAQPage Schema
How do I review an implementation plan for scope creep and unnecessary complexity?▼

You can review an implementation plan for scope creep by running a multi-persona review that assigns specialized AI agents to scrutinize features, complexity, and optimization. It aggregates structured feedback to identify potential issues before development begins.

What is multi-persona plan review and how does it work?▼

Multi-persona plan review is a process where distinct AI personas, such as a complexity-detector and test-coverage-advocate, analyze your software plan. Each persona generates structured concerns and recommendations like CUT or SIMPLIFY for interactive refinement.

How do I use AI agents to check test coverage gaps in my software plan?▼

To check test coverage gaps, you use a specialized test-coverage-advocate persona that evaluates your existing plan. It identifies missing test scenarios and outputs actionable recommendations, allowing you to accept or modify suggestions interactively.

Can I use automated code review techniques during the planning phase before development?▼

Yes, you can apply automated review techniques during planning by using specialized AI personas to analyze your plan for premature optimization and scope management. This prevents over-engineering by generating actionable insights before any code is written.

What is the best way to simplify an over-engineered software implementation plan?▼

The best way to simplify an over-engineered plan is to deploy a feature-trimmer and complexity-detector persona review. They flag unnecessary complexity and output structured recommendations like DEFER TO V2, culminating in an updated and refined plan artifact.

Related Skills

Explore More Software Engineering Skills→
softwaresaltsoftwaresalt
1

plan-review

Validate implementation plans via multi-persona review with P0–P3 severity gating.

Community
Advanced
dralgorhythmdralgorhythm
111

decomposing-tasks

Decompose large tasks into sequenced, testable task sets with dependencies.

Community
Basic
Ramprasad4121Ramprasad4121
1

audit-prep-assistant

Prepare codebases for security review with static analysis and documentation.

Community
Advanced
Nuwanda04Nuwanda04

senior-prompt-engineer

Analyze prompts, evaluate RAG systems, and design agentic workflows.

Community
Advanced