Can it evaluate how well the AI agent performed?

Yes, the Agent Session Evaluation mode analyzes session history to judge output quality, process compliance, and semantic correctness based on universal quality dimensions.

What makes this QA skill different from standard unit tests?

This skill focuses on qualitative assessment and fitness-for-purpose, ensuring the output actually solves the user's problem rather than just passing technical code-level checks.

How do I trigger a quick quality check?

Simply use the /qa command or ask Claude 'is it done' or 'verify work' to invoke a quick verification subagent that checks your work against the task requirements.

Does it support visual or UI testing?

Yes, the Qualitative Assessment mode includes a Visual Analysis Protocol specifically designed for evaluating UI changes and visual artifacts from a user persona perspective.

AcademicOps QA & Verification

Name: AcademicOps QA & Verification
Author: nicsuzor

bynicsuzor

0•

セキュリティとテスト

Provides comprehensive quality assurance, verification, and criteria design to ensure code and agent sessions meet rigorous standards.

AcademicOps QA is a versatile skill for Claude Code designed to manage the entire quality lifecycle, from designing qualitative acceptance criteria and QA plans to executing post-development verification. It goes beyond binary test passes by evaluating fitness-for-purpose through narrative assessment, visual analysis, and agent session performance reviews. Whether you are validating framework integrations, designing QA infrastructure, or performing quick sanity checks before task completion, this skill ensures that work truly aligns with user requirements and operational excellence.

主な機能

010 GitHub stars

02Qualitative assessment focusing on fitness-for-purpose rather than binary checks

03Automated agent session evaluation with trend reporting and insight generation

04Integration validation for structural framework changes and system design

05Built-in support for visual analysis protocols and user persona simulation

06Seven specialized modes covering planning, testing, and performance evaluation

ユースケース

01Running a final verification check to ensure all acceptance criteria are met before completion

02Evaluating previous agent sessions to identify patterns of success or recurring weaknesses

03Creating a detailed QA plan with qualitative dimensions before starting a complex feature

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nicsuzor/academicops qa

For use in Claude.ai and ChatGPT

Download Skill