Memory may preserve causes, but it may not silently become authority.
The CML Lotus Layer defines the human and memory-governance principles for this repository:
- cause before conclusion;
- evidence before durable memory;
- consent before visibility;
- memory without authority;
- context before reuse;
- supersession before erasure;
- human challengeability.
Read the complete bilingual principles in docs/LOTUS.md.
Before accepting or retrieving durable memory, ask:
Does this memory help us understand the cause without taking choice away from the person?
This layer is documentation and human guidance. It is not a runtime component, policy engine, approval mechanism, or autonomous actor, and it grants no execution, delivery, or merge authority.