Show HN: Kage, a framework to maintain and verify Google's OKF agent memory
Kage is a framework that adopts Google's Open Knowledge Format (OKF) for agent memory management. It handles the timing, structure, and verification of memory creation and recall, going beyond Google's standard which does not cover these aspects. Designed for team coordination on shared repositories, Kage works with Claude Code and other coding agents to automate memory maintenance.
Background
- OKF (Open Knowledge Format) is a recently released Google standard for how AI agents store project "memory" — the context, decisions, and facts they learn while coding. It's file-based (plain text in a repo), not a database.
- Kage is a new open-source framework built on OKF. It automates when and how agents save memories, verifies they're accurate and current, and helps agents recall relevant past context. Without Kage, agents using OKF would save memories manually and inconsistently.
- The problem: coding agents (Claude Code, Cursor, etc.) need to remember project-specific context across sessions — past fixes, architecture decisions, config quirks. OKF gives the format; Kage gives the discipline (when to write, what to write, how to keep it fresh).
- Best support for Claude Code via its Hook system. Designed for teams sharing a repo where agent memory must stay consistent across human and AI contributors.