CPR — Conversational Pattern Restoration

Fix robotic AI assistants. Any model. Any provider. Any personality.

Modern LLMs are over-trained toward sterile, corporate communication patterns. CPR identifies the 6 universal humanizing patterns lost during RLHF/fine-tuning and provides a systematic framework to restore them — without triggering sycophancy or hype drift.

Version 4.0: Personality-agnostic + model-size aware + game-theoretically grounded. Works on everything from Haiku to Opus. Small models get heavy scaffolding, large models get a light touch — same voice output regardless of model size. V4 adds mathematical foundations from signaling theory, repeated game analysis, and agency theory that explain why CPR works and catch sycophancy patterns that phrase lists miss.

Quick Start

1. Define your baseline: Use BASELINE_TEMPLATE.md to identify YOUR authentic voice
2. Apply restoration patterns: Read RESTORATION_FRAMEWORK.md — the 6 universal patterns across personality types
3. Prevent drift: Use DRIFT_PREVENTION.md calibrated to YOUR personality
4. Understand the math (optional): Read CPR_V4_GAME_THEORY.md for game theory foundations
5. Reference results: See CROSS_MODEL_RESULTS.md for model-specific notes

What's Included