ki-feedback-design

To improve AI-generated feedback, the skill analyzes existing feedback against evidence-based principles from Shute, Narciss, and Hattie. It identifies pedagogical gaps like vague praise and generates a revised, actionable feedback version tailored to specific student scenarios and learning goals.

Pedagogically effective formative feedback in automated tutoring satisfies evidence-based principles by being specific, timely, and actionable. This skill evaluates AI feedback designs against established frameworks to eliminate generic praise and ensure the text drives meaningful learning progress.

Yes, you can evaluate feedback prompts for K-12 and higher education LMS workflows. The skill applies to design-time prompts, automated tutoring, and LMS feedback workflows, analyzing current designs and providing deployment guidance for integrating the improved feedback text.

You revise vague AI feedback by providing the current design and student scenario as input. The skill identifies gaps like vagueness, applies evidence-based principles, and outputs a revised, actionable feedback message tailored to the specific learning objectives.

This skill applies evidence-based feedback principles by analyzing automated tutoring outputs against frameworks from researchers like Shute, Narciss, and Hattie. It identifies pedagogical shortcomings and generates improved feedback text designed to maximize student learning impact.

Common limitations of generic AI feedback include vagueness and unactionable praise that fail to drive learning progress. This skill identifies these specific pedagogical gaps in current designs and revises the feedback text to align with evidence-based educational principles.