skills.rest
GarethManning
GarethManningCOMMUNITY
Explore all skills by this author →
Website𝕏@worldteachermanGitHub

ai-output-critical-audit-designer

Design student protocols to audit AI text against Ennis's six critical thinking standards.

583|96|Updated Mar 1, 2026
Search Tags:#critical thinking#content evaluation#rubrics#hallucination detection#ai literacy#student assessment#annotation protocol
One-click install
npx skills add https://github.com/GarethManning/education-agent-skills --skill ai-output-critical-audit-designer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-output-critical-audit-designer
Source: https://github.com/GarethManning/education-agent-skills/tree/main/skills/ai-literacy/ai-output-critical-audit-designer
Command: npx skills add https://github.com/GarethManning/education-agent-skills --skill ai-output-critical-audit-designer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of students being misled by confident, fluent AI text by replacing intuitive judgment with a structured protocol to evaluate AI output against Ennis’s six critical thinking standards.

Core Features & Use Cases

  • AI-specific audit protocol: Generates a step-by-step method students can apply to detect AI characteristic failure modes (e.g., overconfident uncertainty-free claims, fabricated precision, shallow depth).
  • Annotation system for close reading: Produces marking codes so students can identify failure types as they read the AI output.
  • Rubric and push-back prompts: Provides Weak/Moderate/Strong descriptors aligned to clarity, accuracy, precision, relevance, depth, and breadth, plus sentence stems that help students challenge AI statements.
  • Teacher modelling script: Includes a think-aloud example showing how an expert auditor applies the standards while explicitly decoupling fluency from reliability.

Quick Start

Use the ai-output-critical-audit-designer to generate an audit and annotation protocol for the AI-generated text about a given subject, at the specified student level, and against the selected Ennis standards.

Frequently Asked Questions about ai-output-critical-audit-designer

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

FAQPage Schema
How do I create a critical thinking rubric to evaluate AI-generated text in the classroom?▼

This Skill generates a structured student-facing protocol to audit AI output against Ennis's six critical thinking standards, producing a Weak/Moderate/Strong scoring rubric aligned to clarity, accuracy, precision, relevance, depth, and breadth.

What is the best way to help students detect AI hallucinations and fabricated precision in essays?▼

Detecting AI hallucinations involves applying a generated annotation protocol that teaches students to mark characteristic AI failure modes like overconfident uncertainty and fabricated precision during close reading of AI explanations.

Can I use this audit protocol design tool for any subject and student year group?▼

Yes, the protocol applies to classroom and tutoring scenarios where learners evaluate AI explanations, arguments, essays, or summaries in any subject and at any year group, adjusted by the provided critical thinking experience level.

How do I teach students to challenge and push back against confident AI writing?▼

You can use the generated push-back sentence stems and a teacher modelling think-aloud script, which demonstrate how an expert auditor decouples AI fluency from reliability to explicitly challenge AI statements.

Does this critical thinking audit tool require any specific software dependencies or components?▼

No, the Skill has no software dependencies or components, requiring only the AI-generated text, the target subject, the student level, and the selected Ennis critical thinking standards to generate the complete audit protocol.

What steps do students follow to annotate and score AI output using Ennis's standards?▼

Students follow stepwise annotation instructions with marking codes to identify failure types, then score the text using the CT-aligned rubric and push-back stems provided in the generated protocol.

Related Skills

Explore More Education & Research Skills→
vvieira010-pixelvvieira010-pixel

literacy-critical-thinking-task-designer

Design critical thinking tasks with evidence evaluation and reasoning prompts.

Community
Advanced
miethemiethe

amcs-validator

Score composed artifacts against a blueprint rubric and output scores, issues, and pass.

Community
Advanced
vvieira010-pixelvvieira010-pixel

ai-output-critical-audit-designer

Designs protocols for auditing AI-generated text against critical thinking standards.

Community
Advanced
dtsongdtsong
1

AI Evaluation

Design evaluation frameworks for AI features with golden datasets and scoring rubrics.

Community
Advanced