Back to Documentation

Creating Prompts

14 steps - 5 minutes

Learn how to create and manage reusable prompts for consistent AI interactions.

What are Prompts?

Prompts are reusable text instructions stored in your library. They can include variables for dynamic content. Prompts are automatically versioned, so you can track changes and revert if needed. Use prompts in chat by inserting them into messages, or in automations by attaching them to agent nodes.

1

Navigate to Prompts

From the sidebar, click on "Prompts" to access your prompt library.

[Screenshot placeholder - Step 1]

2

Click "Create Prompt"

Click the "Create Prompt" button to open the prompt creation form.

[Screenshot placeholder - Step 2]

3

Enter Title

Give your prompt a descriptive title (e.g., "Code Review Template", "Email Draft", "Meeting Summary"). This is what you will see in lists.

[Screenshot placeholder - Step 3]

4

Write the Prompt Text

Enter the prompt content that will be sent to the AI. Be clear and specific about what you want the AI to do. You can use variables like {{variable_name}} for dynamic content.

[Screenshot placeholder - Step 4]

5

Add Description (Optional)

Add a brief description to help you and your team understand when to use this prompt.

[Screenshot placeholder - Step 5]

6

Set Category and Tags

Choose a category (e.g., "general", "coding", "writing") and add tags for better organization and discovery.

[Screenshot placeholder - Step 6]

7

Set Scope

Choose whether this prompt is Personal (only you) or Workspace (your team). You can also make it visible in the community.

[Screenshot placeholder - Step 7]

8

Click "Create"

Save your prompt to make it available for use in chats and automations.

[Screenshot placeholder - Step 8]

9

Using Prompts in Chat

In the chat interface, click the prompts library icon to browse and select prompts. Click a prompt to insert it into your message.

[Screenshot placeholder - Step 9]

10

Using Prompts in Automations

When building automations, drag prompts from the side panel into agent nodes. The prompt will be loaded into the agent configuration.

[Screenshot placeholder - Step 10]

11

Automatic Version Control

Every time you edit a prompt, Uru automatically creates a new version. You can view version history and revert to previous versions if needed.

[Screenshot placeholder - Step 11]

12

Viewing Version History

Click on a prompt, then click "Version History" to see all changes. Each version shows the title, content, and change summary.

[Screenshot placeholder - Step 12]

13

Reverting to a Previous Version

In the version history, click "Revert" next to any version to restore the prompt to that state. This creates a new version.

[Screenshot placeholder - Step 13]

14

Editing Prompts

Click on any prompt in your library to view details, then click "Edit" to modify the title, content, description, category, tags, or scope.

[Screenshot placeholder - Step 14]

Best Practices

  • •Be specific and clear about what you want the AI to do
  • •Use variables like {{name}} for dynamic content that changes each time
  • •Use descriptive titles and add tags for easy discovery
  • •Test prompts in chat before using them in automations
  • •Set workspace scope to share prompts with your team for consistency
  • •Review version history regularly to track how prompts evolve

Next Step

Learn how to create context blocks to provide the AI with relevant information.

Managing Context