prune-review

Audit sprints, features, or codebases for unnecessary complexity and generate a Prune Report.

28|2|Updated Feb 5, 2026
One-click install
npx skills add https://github.com/labs21-dev/agents-stack --skill prune-review
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: prune-review
Source: https://github.com/labs21-dev/agents-stack/tree/main/templates/.agents/skills/using-agents-stack/prune-review
Command: npx skills add https://github.com/labs21-dev/agents-stack --skill prune-review

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

prune-review is a standalone complexity audit specialist dispatched on demand when the user explicitly requests a prune review — on a specific sprint, feature, or codebase-wide. You are not a lifecycle phase. You don't require review.md to exist. You don't produce artifacts that any other phase depends on.

Core Features & Use Cases

  • Executes a structured six-question prune audit against a target sprint, feature, or codebase.
  • Evaluates complexity signals from references/complexity-signals.md and applies scale guidance from references/scale-appropriateness-guide.md to produce a prioritized cut list.
  • Produces a shareable Prune Report that records evidence, recommendations, and escalation paths for stakeholders.

Quick Start

Provide a prune review for the specified sprint or codebase by running through the six PRUNE questions and generating a structured report.

Frequently Asked Questions about prune-review

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

FAQPage Schema
How do I audit unnecessary complexity in my codebase?▼

A complexity audit evaluates your codebase against specific signals and scale appropriateness guidelines to identify unnecessary risk. It applies six structured prune questions to surface and prioritize areas for refactoring or removal.

What is the best way to identify refactor risks in a sprint?▼

The best way to identify refactor risks in a sprint is by applying six prune questions to evaluate codebase signals against scale appropriateness guidelines. This produces a Prune Report with prioritized cut recommendations and escalation paths.

How do I generate a prioritized cut list for a feature review?▼

You generate a prioritized cut list by running an on-demand prune audit against the feature, evaluating complexity signals and scale appropriateness. This yields a Prune Report detailing a defined floor and specific cut recommendations.

Can I run a codebase audit without existing lifecycle artifacts or dependencies?▼

Yes, you can run a standalone codebase audit without existing lifecycle artifacts. This prune review operates independently, requiring no review.md files and producing no downstream artifacts that other lifecycle phases depend on.

When should I request a prune review for my codebase?▼

You should request a prune review when you need on-demand assessment of sprint health or codebase-wide complexity. It is explicitly triggered to surface unnecessary complexity and generate a shareable report with escalation paths.