[Anthropic Reduces Claude Code System Prompts by 80%, New Model Maintains Coding Performance Without Strict Rules] According to Deep Tide TechFlow, on July 25, Anthropic engineers revealed that the team has reduced approximately 80% of the system prompts for the latest Claude model's Claude Code, with no decline in coding evaluation performance. Early versions, due to insufficient model capabilities, included a large number of strict rules in the prompts (such as prohibiting multi-line comments). However, as the new model gained autonomous decision-making abilities, these rules instead caused instruction conflicts and thought divergence. The new strategy shifts to describing objectives (e.g., 'write code that matches the surrounding code') and distributing instructions into tool descriptions and on-demand skill modules. The team explicitly views prompt bloat as technical debt, believing that as model capabilities improve, instructions should be simplified rather than stacked.