What problem does it solve? Long memory files like CLAUDE.md, todo lists, and preference notes consume large amounts of input tokens every time they are loaded into an AI context window. This Skill rewrites them in a terse caveman style that preserves all technical substance while cutting token usage. ## Core Features & Use Cases - Token-reducing compression: Removes articles, filler words, hedging, and redundant phrasing while keeping meaning intact. - Strict preservation rules: Code blocks, inline code, URLs, file paths, commands, technical terms, and markdown structure are never modified. - Safe overwrite with backup: The compressed version overwrites the original file, and a human-readable backup is saved as FILE.original.md. - Validated pipeline: A CLI script detects file type, compresses via Claude, validates output, applies targeted fixes, and retries up to 2 times before reporting failure. - Use Case: Your CLAUDE.md has grown to several thousand tokens of verbose instructions. Run the compression to shrink it dramatically while keeping every command, path, and rule intact. ## Quick Start Ask the AI to compress your memory file by saying: compress memory file /path/to/CLAUDE.md with caveman-compress.
