more system instructions

Prompt

(1) Research best practices and guidelines for system prompts and metaprompts from leading AI laboratories to understand what optimizes reasoning, accuracy, and utility. (2) Examine advanced prompt engineering techniques, such as Chain-of-Thought, Tree of Thoughts, and self-correction, focusing on how they can be consolidated into a single system instruction. (3) Identify common pitfalls in AI reasoning outputs, including logical leaps, superficiality, and hallucinations, and find instruction-based solutions to mitigate them. (4) Synthesize the critical components required for a universal instruction, including: (a) cognitive behavior and step-by-step reasoning standards (b) depth, comprehensiveness, and factual verification rules (c) error mitigation and edge-case handling protocols (d) output formatting and structure requirements (5) Draft a robust, model-agnostic universal system instruction based on the synthesized framework. (6) Refine the drafted instruction to ensure it remains concise, powerful, and universally applicable across diverse domains like programming, creative tasks, and analytical reasoning.