okf-agent-memory
Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclos
What is okf-agent-memory?
Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.
Overview of okf-agent-memory
What it is
OKF Agent Memory is a Git-native persistent memory system for AI coding agents, based on the Open Knowledge Format (OKF) v0.2. It is built in pure Go and requires no external databases or dependencies. The project implements a Dual-Memory Agent Architecture (DMAA) that separates normative working memory (push layer) from semantic domain memory (pull layer) to eliminate prompt bloat.
Key features
- Git-native persistent memory for AI coding agents - Implements Google OKF v0.2 specification - Sub-300µs in-memory BM25 search - Embedded MCP server - Progressive disclosure of knowledge - Dual-Memory Cognitive Architecture (DMAA) separating push and pull layers - Agent Action Grammar (AAG) saving approximately 78–85% tokens compared to natural language instructions - Zero external databases or dependencies - Built in pure Go
Use cases
- Persistent storage of architectural decisions, domain discoveries, and operational facts for AI agents - Reducing context bloat by replacing monolithic prompt files with on-demand retrieval - Avoiding attention drift and RAG blindspots in agent interactions - Maintaining a project knowledge corpus that survives context window resets
Getting started
See the repository README for installation and quick-start instructions.
License and pricing
Licensed under MIT. Open source.
Based on the project's public README and repository metadata.