📚 OKF Agent Memory: Git-Native Memory for Coding Agents

This project tackles a specific pain point for AI coding assistants: how to efficiently manage and retrieve context memory. Instead of bolting on a heavy vector database, it implements a Git-native persistent memory system in pure Go, adhering to Google's OKF v0.2 specification. The technical highlight here is its lightweight and ultra-fast design. It features sub-300µs in-memory BM25 search and an embedded MCP server. By utilizing progressive disclosure, it claims to slash token bloat by 80%. For teams building local coding agents who want to avoid complex database dependencies, this zero-dependency, pure-Go architecture is highly relevant.