trust

What problem does it solve?

It helps you describe and generate the felt experience of trust—especially the asymmetry of how trust is built slowly and broken fast—so responses feel human instead of merely game-theoretic or clinically correct.

Core Features & Use Cases

• Felt-vs-logical framing: contrasts hollow “prediction/probability” explanations with embodied, inside-the-experience explanations.
• Situation-based tests: uses concrete scenarios (like the knife, the confession, the returned wallet) to determine whether an answer truly understands trust as lived risk.
• Failure mode detection: defines what an inauthentic answer sounds like, so the model can self-correct toward the “bridge/knife/hand-on-the-back” texture.

Quick Start

Use the trust skill to rewrite an explanation you are generating so it includes the knife/bridge/hand imagery and a felt consequence analysis for one specific scenario.