Agent: Claude Code, CursorLLM: Claude 3.5, GPT-4#prompt-compression#token-optimization#claude-code#developer-tools#cost-reduction
A macOS menu bar app that optimizes prompts sent to Claude Code and OpenAI Codex, reducing token usage by ~50% through reversible compression of tool output, logs, and boilerplate. Runs locally with zero impact on code workflows, includes savings analytics, and manages an isolated Python runtime for token-saving tools.