skills.rest
giladresisi
giladresisiCOMMUNITY
Explore all skills by this author →
GitHub

system-review

Analyze plan adherence post-implementation to classify divergences and suggest CLAUDE.md updates.

3|Updated Feb 12, 2026
Search Tags:#root cause analysis#ai workflow#process improvement#plan adherence#meta analysis#system review
One-click install
npx skills add https://github.com/giladresisi/ai-dev-env --skill system-review
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: system-review
Source: https://github.com/giladresisi/ai-dev-env/tree/main/skills/system-review
Command: npx skills add https://github.com/giladresisi/ai-dev-env --skill system-review

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill addresses the challenge of identifying and rectifying systemic issues in the AI development workflow, moving beyond code-level fixes to process optimization.

Core Features & Use Cases

  • Plan Adherence Analysis: Evaluates how closely the implemented work followed the original plan.
  • Divergence Classification: Categorizes deviations from the plan as justified or problematic.
  • Root Cause Identification: Pinpoints the underlying reasons for problematic divergences.
  • Process Improvement Suggestions: Recommends updates to CLAUDE.md, plan templates, or new skills to prevent future issues.
  • Use Case: After an AI implements a new feature, this Skill analyzes the execution report and plan to determine if the AI followed the intended architecture, if requirements were misunderstood, or if the planning process itself needs refinement.

Quick Start

Analyze the adherence of the recent implementation to the plan and identify areas for process improvement.

Frequently Asked Questions about system-review

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

FAQPage Schema
How do I analyze AI workflow plan adherence after implementation?▼

Plan adherence analysis evaluates how closely implemented work followed the original plan by comparing generated plans against execution reports to classify divergences and identify systemic workflow issues.

What is root cause analysis for AI development process improvement?▼

Root cause analysis for AI development pinpoints underlying reasons for problematic divergences from plans, moving beyond code-level fixes to recommend systemic workflow enhancements and CLAUDE.md updates.

How do I classify divergences between an implementation plan and execution report?▼

Divergence classification categorizes deviations from the original plan as either justified or problematic by analyzing execution reports and tracing root causes to suggest process improvements.

When do I need a meta-level system review of my AI workflow?▼

A meta-level system review is needed after an AI implements a feature to determine if requirements were misunderstood, if the intended architecture was followed, or if the planning process itself requires refinement.

Can I use process improvement suggestions to update CLAUDE.md and plan templates?▼

Process improvement suggestions recommend specific updates to CLAUDE.md, plan templates, or new skills to prevent future AI workflow issues and rectify systemic development problems.

What artifacts are required for a detailed plan adherence analysis?▼

Plan adherence analysis requires detailed artifacts including PROGRESS.md, generated plans, and execution reports to trace root causes and provide actionable insights for systemic workflow enhancements.

Related Skills

Explore More Software Engineering Skills→
djalmaaraujodjalmaaraujo

linear-issue-create

Create Linear issues from PROGRESS.md plans and update plan files with issue URLs.

Community
Advanced
SummerSecSummerSec
6

skill-optimizer

Analyze SKILL.md frontmatter, workflows, and resources to produce an optimization plan.

Community
Advanced
huangjunsen0406huangjunsen0406
77

trellis-check

Run linting, type-checking, and cross-layer data flow analysis before code commits.

Community
Advanced
dogekikidogekiki

research-review

Review ML research papers through multi-round adversarial critiques.

Community
Advanced