Prompting

Criteria	User Prompts (specific tasks)	System Prompts (behaviour and tone)
Purpose	Task-specific instructions	Overall framework & guideliness
Frequency of Use	Used frequently, often changed	Set once, rarely changed
Scope	Narrow, focused on individual tasks	Broad, applies to all interactions
Content Focus	Specific details, context & desired outcomes	General rules, tone, ethics & brand values
Example	"Write a follow-up email to prospect X about Y product"	"You are a seasoned account executive for a B2B SaaS company..."
Typical Length	Short to medium (1-5 sentences)	Medium to long (paragraph to pages)
Primary Impact	Output content & structure	Overall tone, behavior & approach
When to Use	For each specific task or request	At the beginning of AI solution setup or a new session
Modifiability	Easily modified for each new task	Requires careful consideration to change

Prompt Management

The Prompts interface provides several key features for managing your custom prompts:

• Create: Design new prompts with customizable titles, access levels, and content.

• Share: Share prompts with other users based on configured access permissions.

• Access Control: Set visibility and usage permissions for each prompt.

• Slash Commands: Quickly access prompts using custom slash commands during chat sessions.

Creating and Editing Prompts

When creating or editing a prompt, you can configure the following settings:

• Title: Give your prompt a descriptive name for easy identification.

• Access: Set the access level to control who can view and use the prompt.

• Command: Define a slash command that will trigger the prompt (e.g., /summarize).

• Prompt Content: Write the actual prompt text that will be sent to the model.

Prompt Variables

You can include dynamic prompt variables in your prompts:

• Clipboard Content: Use {{CLIPBOARD}} to insert content from your clipboard.

• Date and Time:

  • {{CURRENT_DATE}}: Current date

  • {{CURRENT_DATETIME}}: Current date and time

  • {{CURRENT_TIME}}: Current time

  • {{CURRENT_TIMEZONE}}: Current timezone

  • {{CURRENT_WEEKDAY}}: Current day of the week

• User Information:

  • {{USER_NAME}}: Current user's name

  • {{USER_LANGUAGE}}: User's selected language

  • {{USER_LOCATION}}: User's location (requires secure connection and permission)

Variable Usage Guidelines

• Enclose variables with double curly braces: {{variable}}

• The {{USER_LOCATION}} variable requires a secure connection and permission.

• The {{CLIPBOARD}} variable requires clipboard access permission from your device.

Access Control and Permissions

Prompt management is controlled by permission settings:

• Prompts Access: Users need the USER_PERMISSIONS_PROMPTS_ACCESS permission to create and manage prompts.

• For detailed information about configuring permissions, refer to the Permissions documentation.

Best Practices

• Use clear, descriptive titles for your prompts

• Create intuitive slash commands that reflect the prompt's purpose

• Document any specific requirements or expected inputs in the prompt description

• Test prompts with different variable combinations to ensure they work as intended

• Consider access levels carefully when sharing prompts with other users, as public sharing makes them available to all users when they use the / command in a chat.

System Prompting refers to the process of designing and providing inputs (prompts) to a system, application, or artificial intelligence (AI) model to elicit a specific response, action, or output. The goal of system prompting is to guide the system towards producing the desired outcome, whether it's generating text, classifying data, making decisions, or executing tasks.

Key Components of System Prompting

1. Prompt Design : Crafting the input (prompt) that will be fed into the system. This involves understanding the system's capabilities, the desired output, and how to effectively communicate with the system.

2. System Understanding : Having a deep understanding of the system's architecture, limitations, and behavior under various inputs.

3. Desired Outcome : Clearly defining what output or action is expected from the system in response to the prompt.

Types of Prompts in System Prompting:

1. Open-Ended Prompts : Encourage free-form responses, often used in generative AI models. Example: "Write a short story about a character who discovers a hidden world."

2. Closed Prompts : Seek specific, constrained responses, commonly used in decision-making systems. Example: "Given a customer's purchase history, recommend a product from the following categories: A, B, or C."

3. Primed Prompts : Include contextual information to influence the system's response. Example: "Considering the current market trends and the company's sustainability goals, outline a strategy for the next quarter."

4. Chain Prompts (or Multi-Step Prompts) : Involve a series of interconnected prompts to achieve a complex outcome. Example: "Plan a trip from New York to Tokyo.

   1. Suggest three flight options.

   2. Recommend accommodations based on the chosen flight.

   3. Provide a 3-day itinerary."

Best Practices for Effective System Prompting

• Clear Definition of Desired Outcomes

• Deep Understanding of the System's Capabilities and Limitations

• Prompt Iteration and Refinement Based on System Responses

• Consideration of Context and Potential Biases

• Continuous Evaluation of Prompt Effectiveness

• Use a New Chat for Each New Topic to Optimize Results

Challenges and Future Directions in System Prompting

• Mitigating Biases in Prompts and System Responses

• Developing Universal Prompting Standards for Interoperability

• Enhancing Explainability of System Responses to Prompts

• Adaptive Prompting Techniques for Dynamic System Environments