plan-review

Diagnose implementation plan estimate variance and generate evidence-based retrospectives.

Updated May 3, 2026
One-click install
npx skills add https://github.com/miethe/boxbrain-2 --skill plan-review-miethe
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: plan-review
Source: https://github.com/miethe/boxbrain-2/tree/main/.claude/skills/plan-review
Command: npx skills add https://github.com/miethe/boxbrain-2 --skill plan-review-miethe

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill reveals why completed implementation plans over- or under-ran their estimates, separating estimation failures from scope changes, blockers, discovery work, and worthwhile quality investments.

Core Features & Use Cases

  • Retrospective Analysis: Compare estimated story points, timelines, phases, files, lines of code, commits, and actual implementation complexity.
  • Heuristic Attribution: Map variance to planning heuristics such as noun counting, dual-implementation overhead, algorithmic complexity, bundled estimates, anchors, and plumbing effort.
  • Evidence-Based Follow-Up: Generate a retrospective with anchor updates, heuristic tuning suggestions, candidate memories, and optional CCDash provenance.
  • Use Case: After a feature is complete, use this Skill to determine whether its estimate missed repository duplication, service complexity, infrastructure plumbing, or execution-related variance.

Quick Start

Use the plan-review skill to create a retrospective for the completed implementation plan at the specified path or feature slug.

Frequently Asked Questions about plan-review

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

FAQPage Schema
How do I run a retrospective on completed implementation plans to analyze estimation variance?

To run a retrospective on completed implementation plans, analyze plan metadata, git history, progress files, and optional CCDash reports. This process diagnoses feature effort and estimation variance by comparing estimated story points against actual implementation complexity.

What is heuristic attribution for story point estimation and how does it work?

Heuristic attribution maps estimation variance to specific planning heuristics like noun counting, dual-implementation overhead, algorithmic complexity, bundled estimates, anchors, and plumbing effort, separating genuine estimation failures from scope changes and blockers.

Why does my feature implementation over-run its story point estimates?

Feature implementations over-run estimates due to estimation failures, scope changes, blockers, discovery work, or quality investments. Analyzing git history and plan metadata reveals whether repository duplication, service complexity, or infrastructure plumbing caused the divergence.

Can I use git history and progress files to generate evidence-based retrospectives?

Yes, you can use git history and progress files to generate evidence-based retrospectives. By performing phase-level variance analysis and scope classification, the retrospective produces anchor updates, heuristic tuning suggestions, and candidate memories.

Do I need CCDash reports to perform a plan review?

You do not need CCDash reports to perform a plan review; they are optional. The analysis can diagnose estimation variance using plan metadata, git history, and progress files, while integrating CCDash reports adds optional provenance to the retrospective.