# Memory, knowledge & artifacts

> The dashboard's read views into what agents have stored and produced — the memory store, the knowledge graph, and rendered artifacts.

Source: https://openap.org/docs/admin-memory

The stores an agent writes to are readable from the console too — what it [remembers](https://openap.org/docs/memory),
the graph it's built, and the [artifacts](https://openap.org/docs/artifacts) it's produced.

## Memory

**Memory** is the read view into the searchable, authorized [memory store](https://openap.org/docs/memory-storage-search) — the kinds
of entry, how many of each, and their scopes. It's the console counterpart of `oap memory`, and it's gated by
the [same per-subject authorization](https://openap.org/docs/memory-authorization) as the store itself.

![Memory — the searchable, authorized memory store: kinds, entry counts, and scopes.](https://openap.org/media/admin-memory.png)

## Knowledge

**Knowledge** surfaces the [knowledge graph](https://openap.org/docs/memory-storage-search) — entities, facts, and communities
(`oap kg`). It's available when the graph provider is configured; without it, the view simply reports that
knowledge isn't wired up on this install.

## Artifacts

**Artifacts** lists the durable, versioned [outputs](https://openap.org/docs/artifacts) agents have rendered — each with its
revisions. It's the read side of `oap artifact`.

![Artifacts — the durable, versioned outputs agents have produced.](https://openap.org/media/admin-artifacts.png)
