BigHugger
GH Repository · okf-memory

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.

stars
652
30-day movement
+12441/day
Related entries
60
Connections
2
agent-appGoopen-knowledge-formatokfmcp-serveragent-memorymakepersistent-memoryknowledge-baseai-agentsbm25git-nativemarkdownknowledge-graphcoding-agentsgomodel-context-protocolgolangdeveloper-toolsllm-memorymcp

okf-agent-memory is a Git-native persistent memory store for AI coding agents, implementing Google's OKF v0.2 format in pure Go. It includes an embedded MCP server, in-memory BM25 search, and progressive disclosure, with no external databases or dependencies.

Reach for it when you want agents to keep searchable, versioned memory in your repo without running a database.

Use it to

  • Give coding agents persistent memory via MCP
  • Store agent knowledge as Git-tracked markdown
  • Search agent memory with BM25
  • Reduce token bloat through progressive disclosure

For Developers building or running AI coding agents

Role
agent-app
Language
Go
Licence
MIT
Forks
46
Open issues
1
Last push
2026-09-14
Latest release
v0.1.0 · 2026-09-05
topicsagent-memorymcpbm25git-nativecoding-agentsgo