> ## Documentation Index
> Fetch the complete documentation index at: https://docs.riggery.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Memory get

> Read one Instance document.

Loads one document from Instance Memory by id or file name.

<h2 id="input">
  Input
</h2>

Inspector **Settings**. Empty-field rules for the Graph: [Previous nodes](/graph/previous-nodes). **Insert value** is on **Document id** and **Document name**.

Need **Document id** or **Document name**. Empty both fails — not previous Result.

On an Agent Tool, **Document id** and **Document name** can be **Agent decides** or **Fixed**. **Max characters** is always your value on the node.

| Field              | Required | Empty           | Notes                                                          |
| ------------------ | -------- | --------------- | -------------------------------------------------------------- |
| **Document id**    | No       | —               | Id from Memory search. Required if **Document name** is empty. |
| **Document name**  | No       | —               | Placeholder `notes.md`. Required if **Document id** is empty.  |
| **Max characters** | No       | Default: `8000` | 500–100000. The model cannot change this.                      |

<h2 id="output">
  Output
</h2>

On the next node, **Previous nodes** lists **`Memory get`: Result** (`{{Memory get.text}}` while the name is unique).

| Output | In menu | Type          | Next node gets                                                                                                                                    |
| ------ | ------- | ------------- | ------------------------------------------------------------------------------------------------------------------------------------------------- |
| Result | Yes     | Markdown text | Header (`## Document: name`, id, mime, size) plus body, clipped to **Max characters**. Missing document: `Document not found in instance memory.` |