> For the complete documentation index, see [llms.txt](https://docs.zenlytic.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.zenlytic.com/legacy/memories.md).

# Memories (legacy)

{% hint style="danger" %}
**Memories are legacy. Do not use them for new context.** Memories are being retired and replaced by [Skills](/using-zenlytic/skills.md). Existing memories will continue to work and will be migrated automatically in a future release, but **for any new context you want to add, do not create a memory**. Use one of these instead:

* **Recurring patterns or domain logic** → create a [Skill](/using-zenlytic/skills.md).
* **Universal rules and terminology** → add to the system prompt.
* **Field- or table-specific guidance** → put it on the view or field as a `zoe_description`. See [Context Surfaces](/core-concepts/context-surfaces.md) for the decision tree.

The rest of this page documents the legacy Memories feature for users who already have memories in their workspace. See [Migrating from Memories and Topics](/migrations/migrating-from-memories-and-topics.md) for a side-by-side comparison.
{% endhint %}

## Memories (legacy)

> Legacy feature. For new context, use [Skills](/using-zenlytic/skills.md), the system prompt, or `zoe_description` on views and fields.

Use memories to teach Zoë about your company, your data preferences, and how you want questions answered. Zoë uses these memories to understand your business and remember your preferences in future responses.

### How memories work

Click "Add to memory" on helpful responses in chat, or create and manage memories in the memory management portal. When you ask Zoë a question, she automatically finds and uses relevant memories to provide more accurate, consistent answers tailored to your business context and preferences.

### Create a memory

#### From chat

Click "Add to memory" under a helpful response to save it.

![Add to memory in chat](/files/GBgnqYYsjqhGgQ7vqBk8)

After saving, you see a confirmation toast that says "Response Added to Memory — Zoë will more consistently pull this data for similar questions in the future." Use "Go to Memory" to open it, or click "Remove from Memory" to undo.

![Response added toast](/files/c7WKILTDrmoK7fjSq4QT)

#### From the memory management portal

Create memories directly in the portal when you want to define behavior outside of a specific chat.

1. Go to Settings → Memory.
2. Click Create new memory, then choose New Text memory or New Explore memory.

![Create a new memory](/files/UVEaKYdKH4lEZzXvK3md)

* **Text memory**: Enter the User question and Zoë response to specify how Zoë should answer similar questions.

![Create a text memory](/files/trJBJEmJ3GQroL05PoTe)

* **Explore memory**: Enter the question, select the time range, metrics, slices, and any filters, then run the question to preview and save.

![Create an explore memory](/files/8FCovQ69o6D4MJSICr3B)

### Manage your memories

View and control all memories in the memory management portal:

1. Click Settings in the left navigation.
2. Open Memory by clicking on the row.
3. Search, edit, or delete memories as needed. Use the pencil icon to edit or the trash icon to delete.

![Editing a memory](/files/kpFh9WPfJWWc2aDcM8Bp)

### What memories can't do

Keep these limitations in mind:

* **Data questions only**: Memories work for data interpretation, not personality changes. Use [custom system prompts](/tips-and-tricks/zoe_context_ingestion.md) to modify Zoë's behavior or personality.
* **Opt-in only**: Memories save only what you explicitly add or create. They do not learn from negative feedback. Use [custom system prompts](/tips-and-tricks/zoe_context_ingestion.md) to provide general guidance about tone or style.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.zenlytic.com/legacy/memories.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.