OpenClaw 记忆分配策略

一份关于 OpenClaw 记忆结构的实用指南:该记什么、存在哪里,以及记忆如何改变工作区质量。

Memory is one of the biggest differences between a disposable assistant and a workspace that feels trained over time.

Useful places to explore first:

What memory should hold

Good OpenClaw memory usually stores:

  • user preferences
  • project constraints
  • architectural decisions
  • recurring operating rules
  • unresolved questions

What memory should not hold

Avoid storing:

  • transient tool noise
  • huge duplicate notes
  • stale snapshots that no one updates

Practical structures

Common patterns include:

  • a decision log
  • a project ledger
  • role-specific notes
  • a small set of stable operating principles

Study these examples:

Why memory matters

Without memory, workspaces tend to rediscover the same context over and over. With memory, they become easier to guide and easier to trust.

相关路径