
Token-efficient MCP server for code agent context, memory, and Bash output.
Visit Token SaviorToken Savior is an open-source MCP server designed to optimize token usage for coding AI agents, notably Claude Code, Cursor, Gemini, and Codex. It reduces repetitive file reading, introduces persistent memory for session continuity, and aggressively compacts noisy Bash outputs, providing massive token and time savings when navigating large codebases. Ideal for developers and teams leveraging AI coding agents, Token Savior integrates structural code navigation and a sandboxed memory engine via a Python package and CLI tools, and operates as a plug-and-play Model Context Protocol (MCP) component.
Visit Token Savior's official website for product details and getting started.
Comprehensive guides and API references for Token Savior.