inspect

Audit a plan against implemented code and generate VERDICT.md with dimension scores.

6|3|Updated Feb 11, 2026

Search Tags:#code-review #inspect #verdict #agent-teams #plan-audit #risk-context

Ready to run

npx skills add https://github.com/vinhnxv/rune --skill inspect-vinhnxv

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: inspect
Source: https://github.com/vinhnxv/rune/tree/main/plugins/rune/skills/inspect
Command: npx skills add https://github.com/vinhnxv/rune --skill inspect-vinhnxv

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Plan-vs-implementation audit and verification across a multi-agent setup, ensuring specification compliance and uncovering gaps between plan and code.

Core Features & Use Cases

Multi-agent plan audit with four built-in inspector Ashes: grace-warden, ruin-prophet, sight-oracle, vigil-keeper.
Generates VERDICT.md containing requirement matrix, dimension scores, gap analysis, and actionable recommendations.
Risk-context injection and phase-based orchestration to verify completeness, security, and design fidelity.

Quick Start

Provide a plan (file or inline) to trigger a full inspector run and generate VERDICT.md.

Frequently Asked Questions about inspect

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

FAQPage Schema

How do I audit implemented code against a project plan?▼

To audit implemented code against a project plan, provide the plan file or inline description to trigger a multi-agent inspection. This evaluates 10 dimensions including correctness, security, and performance, generating a VERDICT.md with a requirement matrix, gap analysis, and actionable recommendations.

What is a plan-vs-implementation audit?▼

A plan-vs-implementation audit verifies specification compliance by identifying gaps between a project plan and its actual code. It uses multi-agent inspection across 10 dimensions like design fidelity and observability to produce a requirement matrix and gap analysis.

Can I use a multi-agent setup to verify specification compliance?▼

Yes, you can verify specification compliance using a multi-agent setup with four built-in inspectors. These agents apply phase-based orchestration and risk-context injection to evaluate completeness, security, and failure modes, outputting actionable recommendations in a VERDICT.md file.

How do I generate a requirement matrix and gap analysis for my code?▼

You generate a requirement matrix and gap analysis by running a multi-agent plan audit on your plan file or inline description. The inspection evaluates 10 dimensions like maintainability and tests, producing a VERDICT.md file containing the matrix, dimension scores, and actionable recommendations.

What's the best way to evaluate failure modes and design fidelity in a project?▼

The best way to evaluate failure modes and design fidelity is through a multi-agent plan audit that applies risk-context injection. This process assesses 10 dimensions of your implemented code against the plan, yielding dimension scores and actionable recommendations in a VERDICT.md file.