A Model Context Protocol server that wraps multiple backend servers for token-efficient tool discovery via lazy loading. It enables AI models to browse available servers and fetch specific tool schemas on-demand, significantly reducing initial context overhead.
View & run on Rokha →