> For the complete documentation index, see [llms.txt](https://docs.katara.ai/about/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.katara.ai/about/platform/shared-memory-layer.md).

# Shared Memory Layer

The Shared Memory Layer is Katara's governed memory layer. It stores context agents need across sessions, workflows, and approved use cases while keeping scope, retention, and reuse under policy.

For engineers, it is persistent state with guardrails. For governance teams, it is the part of the platform that keeps AI continuity useful without making recall broad, unpredictable, or overexposed.

## What it governs

The Shared Memory Layer is documented as a way to:

* preserve useful workflow context across sessions;
* scope memory by policy, user, application, or workflow;
* redact or expire context that should not persist;
* keep AI continuity understandable and auditable.

## How to think about it

Memory is useful only when it is governed. Katara decides what context is eligible to persist, what must be redacted, and when stored context should expire or stay out of future use.

Detailed retention configuration is intentionally not covered here.