do-system-review

Review agent system workflows and categorize design risks and failures.

Updated Feb 17, 2026
One-click install
npx skills add https://github.com/purefunctions/my-agent-config --skill do-system-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: do-system-review
Source: https://github.com/purefunctions/my-agent-config/tree/main/skills/skills/do-system-review
Command: npx skills add https://github.com/purefunctions/my-agent-config --skill do-system-review

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the need for a thorough and systematic review of agent system workflows, identifying areas for improvement and providing actionable recommendations.

Core Features & Use Cases

  • Systematic Review: Analyze implementation, plan adherence, deviations, maintainability, and human corrections.
  • Risk Tiers: Categorize design risks into tiers for appropriate handling.
  • Failure Classification: Classify failures into various categories for targeted fixes.
  • Noise Filter: Recommend changes based on evidence, prevention, actionability, and impact.
  • Output Format: Provide structured output for review and improvement tracking.

Quick Start

Run the 'do-system-review' skill with the required artifacts to initiate a system review.

Frequently Asked Questions about do-system-review

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

FAQPage Schema
How do I conduct a comprehensive system review for agent workflows?

A system review for agent workflows analyzes implementation, plan adherence, deviations, and maintainability to identify areas for improvement and provide actionable recommendations.

What is the best way to categorize design risks in an agent workflow review?

The best way to categorize design risks is by assigning them into specific risk tiers, which allows for appropriate handling and prioritized fixes during the system review process.

How do I classify failures in agent workflows for targeted fixes?

You classify failures in agent workflows by grouping them into various failure categories, enabling targeted fixes and preventing similar issues in future system designs.

Can I filter out noise when providing improvement recommendations for agent systems?

Yes, you can filter noise by recommending changes based on evidence, prevention, actionability, and impact, ensuring that only meaningful improvements are tracked.

Does the system review output format support improvement tracking?

Yes, the system review provides a structured output format specifically designed to facilitate review processes and enable continuous improvement tracking for agent workflows.