mastermind-standard

To prevent AI performance drift in long-running Claude sessions, you need an internal self-assessment loop that enforces a performance ceiling. This skill applies a Mastermind Gate to check depth, context, and mission alignment before every response to sustain elite output standards.

A Mastermind Gate for maintaining AI output consistency is an internal self-assessment mechanism that checks ceiling, depth, context, and mission alignment before generating every response. It ensures Claude adheres to user-stated goals through literal-read discipline and correction reception.

To maintain consistent output quality across multi-domain agentic workflows, implement session persistence and decay-closure protocols. This sustains rigorous performance on high-stakes tasks by enforcing persistent state and applying literal-read discipline across all domains.

Yes, you can use session persistence to enforce standards for high-stakes tasks. By applying decay-closure protocols and persistent state, the skill prevents Claude from drifting to baseline performance during extended, complex operations, ensuring consistent rigor.

You need a self-assessment loop for AI performance when executing long-running, high-stakes, multi-domain tasks where consistency and rigor are critical. It prevents performance drift by applying a Mastermind Gate to verify mission alignment and context depth before every output.

To stop Claude from dropping to baseline performance over time, enforce an internal standard with persistent state across sessions. This skill implements Pillars 1-7, including correction reception and literal-read discipline, to actively prevent degradation and maintain the ceiling.