probe-understanding-professor

Quizzes users with adaptive professor-style questions and writes scored markdown understanding reports.

3|Updated Feb 15, 2025
One-click install
npx skills add https://github.com/mark-torres10/ai_tools --skill probe-understanding-professor-mark-torres10
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: probe-understanding-professor
Source: https://github.com/mark-torres10/ai_tools/tree/main/skills/probe-understanding-professor
Command: npx skills add https://github.com/mark-torres10/ai_tools --skill probe-understanding-professor-mark-torres10

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? It verifies whether you truly understand a technical topic by testing mechanism-level reasoning instead of accepting surface-level explanations. ## Core Features & Use Cases - Adaptive Questioning Loop: Asks one question at a time, raising or lowering difficulty based on each answer's quality. - Multi-Level Explanation Testing: Requires explaining topics intuitively, precisely, and formally, plus comparison and transfer to new problems. - Scored Markdown Report: Writes a rubric-based scorecard (10 categories out of 10) with critique, an improved answer, and practice questions to a timestamped directory. - Use Case: Ask to be quizzed on backpropagation; the professor probes fundamentals, mechanism, and edge cases, then saves a full evaluation report. ## Quick Start Quiz me on transformer attention and increase the difficulty as I answer, then save the scored report.

Frequently Asked Questions about probe-understanding-professor

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

FAQPage Schema
How do I get quizzed on a technical topic by an AI?▼

Start a session by naming the topic, your current familiarity, and target depth. The professor asks one question at a time, critiques each answer, and adjusts difficulty based on your performance.

How does adaptive difficulty work in a quiz session?▼

Difficulty adjusts after every answer. Weak answers trigger narrower fundamental questions, solid answers raise abstraction and novelty, and excellent answers lead to teach-back, contrast, or transfer challenges.

What topics can the understanding probe cover?▼

It covers any technical topic including engineering, mathematics, AI, and systems concepts. The questioning focuses on mechanism, comparison, transfer, assumptions, and failure cases rather than topic-specific trivia.

Where is the quiz evaluation report saved?▼

The report is written as markdown to probe_understanding/<timestamp>_<topic_slug>/main.md. It includes session metadata, ten rubric scores out of 10, gaps, an improved answer, and practice questions.

What scoring rubric is used to evaluate understanding?▼

Ten categories are scored 0-10: comprehension, correctness, precision, depth, reasoning, transfer, edge cases, communication, brevity, and confidence calibration. Scores are conservative, with 8-10 reserved for rigorous, defensible answers.