Agent long-term memory infrastructure: a desensitization → distillation → reconciliation and writing pipeline, with three memory layers and three integration methods: MCP, LangGraph, and Skill
1 file for Codex and OpenCode: agent-memory AGENTS.md — 3,997 tokens loaded in every session.
AGENTS.md A 3,997 tok These files are ac0033/agent-memory's own configuration — they tell Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.