gsd-eval-review

Audit executed AI phase evaluation coverage against AI-SPEC.md and generate an EVAL-REVIEW.md remediation plan.

Updated May 21, 2026
One-click install
npx skills add https://github.com/jedmamosto/m-and-ms --skill gsd-eval-review-jedmamosto
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gsd-eval-review
Source: https://github.com/jedmamosto/m-and-ms/tree/main/.agents/skills/gsd-eval-review
Command: npx skills add https://github.com/jedmamosto/m-and-ms --skill gsd-eval-review-jedmamosto

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It resolves the risk of incomplete or misapplied AI evaluation coverage by checking whether an executed AI phase actually implemented the evaluation strategy defined in AI-SPEC.md.

Core Features & Use Cases

  • Evaluation coverage audit: Verifies whether required evaluation gates and coverage from AI-SPEC.md were implemented for a completed phase.
  • Actionable remediation plan: Produces an EVAL-REVIEW.md file that includes a score, verdict, identified gaps, and specific next steps to remediate.
  • Phase-scoped retro review: Supports reviewing the last completed phase by default or targeting a specific phase number when provided.

Quick Start

Run gsd-eval-review to generate EVAL-REVIEW.md for the last completed AI phase.

Frequently Asked Questions about gsd-eval-review

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

FAQPage Schema
How do I audit AI evaluation coverage against a declared evaluation strategy?

AI evaluation coverage auditing compares executed AI phase evaluation gates against the declared evaluation strategy in AI-SPEC.md, identifying gaps between spec and implementation to ensure workflow verification completeness.

How do I generate a remediation plan for incomplete AI workflow evaluation gates?

Generating a remediation plan produces an EVAL-REVIEW.md file containing an evaluation score, verdict, identified phase coverage gaps, and specific next steps to remediate incomplete AI workflow evaluation gates.

Can I review evaluation coverage for a specific completed AI phase instead of the last one?

Phase-scoped retro review supports targeting a specific completed phase number via a provided argument, defaulting to the last completed AI phase to verify evaluation gate coverage end-to-end.

What is phase-by-phase evaluation gate verification for completed AI workflows?

Phase-by-phase evaluation gate verification is a retroactive review process that checks whether an executed AI phase actually implemented the evaluation strategy defined in AI-SPEC.md, preserving workflow gates end-to-end.

Does AI evaluation coverage auditing require preserving workflow gates end-to-end?

AI evaluation coverage auditing requires preserving workflow gates end-to-end and reading the referenced evaluation materials to accurately verify phase coverage against the declared evaluation strategy.

What's the best way to identify gaps between an evaluation spec and its implementation?

Identifying spec-to-implementation gaps requires reading the referenced evaluation materials in AI-SPEC.md, comparing declared evaluation strategy against executed phase coverage, and outputting a scored remediation plan.