plan-audit

Audit the active plan and generate a Spec Impact section.

Updated Mar 10, 2026
One-click install
npx skills add https://github.com/iamsteveholmes/momentum --skill plan-audit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: plan-audit
Source: https://github.com/iamsteveholmes/momentum/tree/main/skills/momentum/skills/plan-audit
Command: npx skills add https://github.com/iamsteveholmes/momentum --skill plan-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Audits the active plan to assess its spec impact, classifies it as trivial or substantive, and writes a Spec Impact section to unblock plan exit.

Core Features & Use Cases

  • Classifies plans using guidance from the Spec Capture Guide to determine the scope of changes and potential impacts.
  • Creates a process story for substantive plans, and performs targeted spec audits to ensure alignment with architecture and epic goals.
  • Writes a consolidated Spec Impact section to the plan file and coordinates AVFL-based validation and Go/No-Go decisions.

Quick Start

Install and run the plan-audit workflow against the active plan to trigger the Spec Impact generation and process-story creation.

Frequently Asked Questions about plan-audit

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

FAQPage Schema
How do I audit a plan for spec impact before a go/no-go decision?

To audit plan spec impact, the active plan is loaded from ~/.claude/plans and classified as trivial or substantive using spec-capture guidance to determine go/no-go readiness. Substantive plans trigger a process story creation and targeted spec section audits to ensure alignment with architecture and epic goals.

What is a process story in plan auditing and when is it needed?

A process story in plan auditing is a narrative record generated for substantive plans to capture the execution flow and spec alignment context. It is needed when a plan is classified as substantive rather than trivial, ensuring architecture and epic goals are verified before proceeding.

How do I classify a development plan as trivial or substantive?

Classifying a plan as trivial or substantive relies on signals defined in the spec-capture-guide reference to evaluate the scope of changes and potential impacts. Substantive classifications require deeper spec audits and AVFL validation checkpoints before a go/no-go decision is finalized.

Can AVFL checkpoints be run automatically during plan exit validation?

AVFL checkpoints can be run during plan exit validation when applicable, coordinating validation alongside the generated Spec Impact section. The workflow updates sprint-status and plan files with the final go/no-go decision after these checkpoints pass.

Does plan auditing work without external dependencies?

Plan auditing works without external dependencies, relying entirely on internal scripts and reference guides like spec-capture-guide.md. It operates by locating the active plan locally and extracting the title from the first H1 heading to begin the classification and audit process.

What are the limitations of automated spec impact generation for plans?

Automated spec impact generation is limited to plans located in the ~/.claude/plans directory and requires an H1 heading to extract the plan title. Plans lacking proper heading structure or stored outside this directory cannot be audited or classified by the workflow.