# Managing the Structural Agent and Agent chats
## **How chats are stored**
Tonic Structural stores chats in browser storage.
## **Minimizing and displaying the chat**
To minimize the Agent chat, click the minimize icon.
Minimize option for the Structural Agent
To restore a minimized chat, click the Agent icon.
Minimized Agent
## **Starting a new chat**
To clear the current chat and start a new one, click the delete icon in the Agent heading.
Delete icon to clear the Agent chat
Clearing the chat also clears the browser storage for that chat.
## **Undoing an action**
Before it performs an action to modify the workspace, the Agent takes snapshots of the current workspace configuration.
To undo an action, or to revert to a previous point from the current conversation, prompt the Agent.
For example:
* `Undo that action.`
* `Revert the generator assignments for the datetime columns.`
* `Go back to before we assigned generators to the name columns.`
If needed, the Agent can ask you to select a snapshot to revert to.
## **Copying the most recent response from the Agent**
A response from the Agent might be useful elsewhere. For example, you can prompt the Agent to generate Structural API calls, or the Agent might return a useful summary of the data and its status.
To copy the most recent Agent response, click the copy icon.
Copy response icon for the Structural Agent
## **Providing feedback to the Agent**
You can provide basic positive or negative feedback to indicate the helpfulness of a response. The Agent then uses that feedback to adjust its future responses in the current chat.
To provide positive feedback, click the thumbs up icon.
To provide negative feedback, click the thumbs down icon.
---
# Agent Instructions: 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.tonic.ai/app/structural-agent/managing-the-structural-agent-and-agent-chats.md?ask=
```
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.
