Which files does this skill primarily manage?

The skill focuses on maintaining docs/04_evaluation/EVALUATION_PLAN.md and cross-referencing requirements found in docs/01_requirements/NFR.en.md.

How does this skill interact with Non-Functional Requirements (NFRs)?

It links defined metrics directly to NFRs, ensuring that high-level technical goals like performance or reliability are validated through specific, measurable data points.

What is the primary goal of the Evaluation Plan skill?

It ensures that project quality and progress are measurable by defining specific metrics, thresholds, and measurement methods within your documentation.

When is the best time to apply this skill?

Apply this skill whenever you are defining new metrics, adjusting acceptance criteria, or refining the methods used to measure software quality.

AI Evaluation & Metrics Planner

Name: AI Evaluation & Metrics Planner
Author: gw-ai-security

bygw-ai-security

0•

보안 및 테스팅

Manages evaluation metrics, thresholds, and measurement methods to ensure project progress and quality are quantifiable and documentable.

The Evaluation Plan skill provides a structured framework for defining and maintaining the quantitative benchmarks of a software project. It helps teams move beyond subjective feedback by standardizing how metrics, measurement protocols, and acceptance criteria are documented. By linking these metrics to non-functional requirements (NFRs), the skill ensures that performance and quality goals are always grounded in testable data. It is particularly useful for AI-driven projects, like the ATS CV Scorer, where clear success thresholds and rigorous evaluation methods are critical for proving model efficacy.

주요 기능

01Standardize quantitative evaluation metrics and success thresholds

02Map technical metrics directly to non-functional requirements (NFRs)

030 GitHub stars

04Validate that all project statements remain testable and objective

05Define and update standardized measurement methods for quality assurance

06Maintain automated synchronization with EVALUATION_PLAN.md documentation

사용 사례

01Aligning project deliverables with core non-functional requirements

02Updating measurement methods when transitioning from manual to automated validation

03Defining new performance thresholds for an AI scoring engine

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add gw-ai-security/ats-cv-scorer evaluation-plan

For use in Claude.ai and ChatGPT

Download Skill