Optimizes documentation and prompts for maximum LLM token efficiency and improved instruction following.
The text-optimize skill is a specialized utility designed to maximize Claude's context window efficiency by compressing verbose documentation, project rules, and prompts without losing critical information. It offers three distinct modes—Light, Medium, and Deep—allowing users to choose between simple text cleanup and aggressive structural rephrasing. By applying a rigorous set of rules focused on token efficiency, LLM perception, and reference integrity, it ensures that your project's CLAUDE.md, agent descriptions, and technical guides are both concise and highly effective for AI consumption.
主要功能
01Claude-specific behavioral tuning to improve prompt adherence and prevent overengineering.
02Integrated reference verification for file paths, URLs, and circular dependencies.
03Automated transformation of prose into token-efficient tables, bullets, and one-liners.
04Three-tiered optimization modes (Light, Medium, Deep) for tailored content compression.
