plan-review

Coordinates parallel expert reviewers to evaluate implementation plans and synthesize structured feedback.

Updated Mar 8, 2026
One-click install
npx skills add https://github.com/ambroselittle/agent-skills --skill plan-review-ambroselittle
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: plan-review
Source: https://github.com/ambroselittle/agent-skills/tree/main/skills/plan-review
Command: npx skills add https://github.com/ambroselittle/agent-skills --skill plan-review-ambroselittle

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Plan Review skill coordinates parallel expert reviewers to evaluate a planned implementation for architectural soundness, completeness, and feasibility, then synthesizes findings into actionable feedback for decision-makers.

Core Features & Use Cases

  • Parallel reviewer orchestration: spawn architecture, completeness, security, and testing reviewers in parallel to speed evaluation.
  • Plan validation and synthesis: read the plan, collect findings in standardized MISMATCHES, IMPROVEMENTS, and ALTERNATIVES, and store results for traceability.
  • Context-aware review: load repository conventions from CLAUDE.md when present and apply repo-specific reviewers and conventions.
  • Use cases: accelerate early-stage plan validation between teammates, and provide structured feedback before implementation begins.

Quick Start

Run the plan-review workflow after you generate a plan with /plan-work to start the parallel assessment.

Frequently Asked Questions about plan-review

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

FAQPage Schema
How do I get structured feedback on an implementation plan before writing code?

To get structured feedback on an implementation plan, you can run parallel expert reviews that evaluate architectural soundness, completeness, and feasibility, synthesizing findings into actionable categories like mismatches and improvements for decision-makers.

Can I automate architecture and security reviews to run in parallel?

Yes, you can automate architecture, security, and completeness reviews to run in parallel. This orchestration speeds up early-stage plan validation by spawning multiple expert reviewers simultaneously to evaluate the planned implementation.

How does parallel plan validation work with repository conventions?

Parallel plan validation works by loading repository conventions from CLAUDE.md when present. It applies these repo-specific conventions and reviewers during the assessment to ensure the planned implementation aligns with existing project standards.

What is the best way to organize plan review findings for traceability?

The best way to organize plan review findings for traceability is to enforce a structured reviewer format using MISMATCHES, IMPROVEMENTS, and ALTERNATIVES. Findings are saved and presented category-by-category before proceeding to the next step.

How do I locate a plan file for review using a work slug or branch?

To locate a plan file for review, the system accepts a work slug, a full path, or auto-discovers the file from the current Git branch, enabling flexible review workflows without requiring manual file path specification.

Does plan review work without manually specifying the file path?

Yes, plan review works without manual file paths by auto-discovering the plan file from the current branch. This enables flexible review workflows when you have a planned implementation ready for validation.