What problem does it solve? Moderators unintentionally bias usability tests by rescuing struggling participants, asking leading questions, or explaining the design, which destroys the behavioral evidence the study exists to capture. This Skill provides the exact scripts, probe techniques, and decision rules that keep the participant's behavior — not the moderator's influence — producing the data. ## Core Features & Use Cases - Think-aloud intro script and session arc: A word-for-word facilitation script covering consent, rapport, narration demo, task loop, wrap-up, and observer debrief. - Mid-task decision tree and neutral probe phrasebook: Exact responses for every common participant behavior (going quiet, asking for help, self-blaming, wandering off-task) using echo, boomerang, and Columbo probes, plus leading-to-neutral question rewrites. - Rescue rule and atomic note capture: A five-step intervention tree with a smallest-first hint ladder, plus timestamped one-observation-per-note logging with per-task outcome codes that feed directly into analysis. - Use Case: You are about to run a moderated remote test of a booking prototype with five participants and observers from the product team. Use this Skill to brief observers, deliver the think-aloud intro, handle a participant who gets stuck on task 2 without rescuing them, and produce clean atomic notes for the post-study analysis. ## Quick Start Help me moderate a live usability test session with a real participant, including the think-aloud intro script and what to say when they get stuck.
