gsd-eval-review

Audit AI phase evaluation coverage and produce a remediation plan in YAML frontmatter.

Updated Apr 22, 2025

Search Tags:#AI workflow #AI evaluation #remediation plan #evaluation coverage #AI phase

Ready to run

npx skills add https://github.com/mcampbellr/.dotfiles --skill gsd-eval-review-mcampbellr

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: gsd-eval-review
Source: https://github.com/mcampbellr/.dotfiles/tree/main/claude/.claude/skills/gsd-eval-review
Command: npx skills add https://github.com/mcampbellr/.dotfiles --skill gsd-eval-review-mcampbellr

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill automates the process of evaluating the coverage of an executed AI phase, ensuring that the evaluation strategy from AI-SPEC.md is implemented and producing a remediation plan.

Core Features & Use Cases

Evaluation Coverage Audit: Checks whether the evaluation strategy from AI-SPEC.md was followed.
Remediation Plan Production: Outputs an EVAL-REVIEW.md file with a score, verdict, gaps, and a remediation plan.
Use Case: After an AI phase is completed, run this skill to get a detailed evaluation report and a plan for any necessary improvements.

Quick Start

Run the gsd-eval-review skill with the phase number, e.g., 'gsd-eval-review 2'.

Frequently Asked Questions about gsd-eval-review

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

FAQPage Schema

How do I generate a remediation plan for AI workflow evaluation gaps?▼

To generate a remediation plan for AI workflow evaluation gaps, run an audit of the executed AI phase against your AI-SPEC.md evaluation strategy. The process outputs an EVAL-REVIEW.md file containing a coverage score, verdict, identified gaps, and a detailed remediation plan.

What is AI phase evaluation coverage and when do I need to audit it?▼

AI phase evaluation coverage measures whether the evaluation strategy defined in AI-SPEC.md was fully implemented during an AI workflow. You need to audit this coverage immediately after an AI phase is completed to verify compliance and identify any evaluation gaps before moving forward.

How do I check if my AI workflow evaluation strategy from AI-SPEC.md was followed?▼

You can check if your AI workflow evaluation strategy from AI-SPEC.md was followed by running an evaluation coverage audit on the completed phase. This audit cross-references your workflow execution data with AI-SPEC.md requirements to produce a compliance verdict and highlight missed evaluation criteria.

Do I need specific reference materials to audit AI workflow evaluation coverage?▼

Yes, auditing AI workflow evaluation coverage requires access to evaluation reference materials and workflow execution data. The audit process uses YAML frontmatter and applies these reference inputs to ensure the executed phase complies with the evaluation strategy defined in AI-SPEC.md.

What is the best way to document AI phase evaluation gaps and remediation steps?▼

The best way to document AI phase evaluation gaps and remediation steps is to output a structured EVAL-REVIEW.md file. This file captures the evaluation coverage score, final verdict, specific gaps in the evaluation strategy, and a actionable remediation plan for necessary improvements.

Why does my AI phase evaluation coverage audit show gaps after execution?▼

Your AI phase evaluation coverage audit shows gaps when the executed workflow did not fully implement the evaluation strategy defined in AI-SPEC.md. The audit captures these deviations in an EVAL-REVIEW.md file, providing a score and a targeted remediation plan to resolve the missing evaluation criteria.