codebase-memory-mcp is a code intelligence engine exposed as a Model Context Protocol (MCP) server that gives AI coding agents fast, structural understanding of a codebase. It full-indexes an average repository in milliseconds and the Linux kernel (28M lines, 75K files) in about three minutes, answering structural queries in under one millisecond. Parsing is done through tree-sitter AST analysis across 158 languages, enhanced with a Hybrid LSP layer for semantic accuracy in 9 languages. It ships as a single static binary with zero dependencies for macOS, Linux, and Windows, building a persistent knowledge graph that agents can query as memory.