PromptDecipher: Supporting AI Tutor Authoring Through Editable Simulated Interactions

Key Finding

Teachers virtually never test AI tutoring bots before student deployment; PromptDecipher enforces QA as a first-class activity by letting teachers edit bot responses directly.

Synthesis

PromptDecipher addresses a critical gap in AI tutor deployment: teacher quality assurance. A formative study revealed that educators authoring AI tutoring chatbots virtually never systematically test them before student deployment — a finding with serious implications for AI Tutor Safety Harms and educational quality. The system shifts the authoring paradigm from abstract prompt writing to direct correction-based interaction: teachers edit undesirable bot responses in a live chat preview, and an automated pipeline analyzes the correction, proposes a system prompt rewrite, and validates across test scenarios. This bridges the Teacher Role gap between classroom practitioner and AI system designer — a tension also explored in AI TPACK Teacher Multi Agent Workflow, which found that effective AI integration requires systems thinking beyond simple tool use. PromptDecipher's QA enforcement resonates with the Agentic Workflows Education paradigm of using AI to scaffold human roles. By embedding testing into the authoring workflow, the system also mitigates the kind of diagnostic failures identified in LLM Tutoring Feedback Diagnosis Gap, where LLMs struggle precisely where feedback matters most.

Connected Concepts

Connected Articles

Citation

J, A.K.M.X.R.S. (2026). PromptDecipher: Supporting AI Tutor Authoring Through Editable Simulated Interactions. practice, however, teachers rarely fulfill these roles