sd-improve

Coordinate multi-perspective project and package reviews to surface improvement points.

2|2|Updated Jul 10, 2024
One-click install
npx skills add https://github.com/kslhunter/simplysm --skill sd-improve
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sd-improve
Source: https://github.com/kslhunter/simplysm/tree/main/.back/260203/sd-improve
Command: npx skills add https://github.com/kslhunter/simplysm --skill sd-improve

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinate comprehensive reviews of project and package problems to surface improvement points.

Core Features & Use Cases

  • Orchestrates parallel evaluation across logic, code, architect, test, and docs perspectives.
  • Merges and validates findings, then surfaces prioritized recommendations for remediation.
  • Use Case: Run against a whole monorepo to identify cross-cutting improvements, or target a single package to focus on its quality hotspots.

Quick Start

Invoke sd-improve with an optional package name to review the entire project or a specific package directory.

Frequently Asked Questions about sd-improve

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

FAQPage Schema
How do I automate a comprehensive architecture review for my project?

To automate a comprehensive architecture review, coordinate parallel scans across logic, code, tests, and docs perspectives to surface improvement points. This orchestrates multiple evaluations simultaneously, merging and validating findings into prioritized, actionable recommendations.

What is the best way to analyze code quality across a monorepo?

Analyzing code quality across a monorepo is best done by running a coordinated project analysis that identifies cross-cutting improvements. It applies parallel evaluation over the entire project, merging results to deliver structured reporting on quality hotspots.

How do I run a targeted package review for specific code quality issues?

To run a targeted package review, invoke the analysis with a specific package name to focus on that directory. This narrows the comprehensive logic, code, architecture, test, and docs evaluation to surface quality hotspots within that isolated scope.

Can I include human-in-the-loop decisions during an automated project analysis?

Yes, you can include human-in-the-loop decisions during an automated project analysis. The workflow supports optional human checkpoints after parallel scans and merged validation, allowing manual decisions before delivering structured, actionable feedback.

How does a multi-step workflow validate documentation and test improvements?

A multi-step workflow validates documentation and test improvements by running parallel scans from dedicated perspectives, then merging and validating the findings. This process consolidates results into grouped reporting, ensuring structured feedback for your project.

Do I need specific dependencies to orchestrate project improvement reviews?

No specific dependencies are required to orchestrate project improvement reviews. The workflow applies parallel scans across architecture, logic, tests, and docs natively, merging and validating findings to deliver structured feedback without external packages.