Framework-Driven AI Prompt Generator

Task: Defines the specific task or tasks that the AI is expected to perform, defining "what" needs to be done, like generate blog post, answering questions, providing explanations, etc.

Action: Outlines "how" the AI accomplishes the defined task, detailing its steps or behaviors. For example, it might involve synthesizing information to generate a summary, analyzing data to provide insights, or engaging in a dialogue to answer questions.

Goal: Defines the desired outcome or purpose of the interaction, answering "why" the task is being undertaken. For example, the goal might be to educate users, entertain them, solve a specific problem, or generate a particular type of content.

By systematically defining tasks, actions, and goals, the TAG framework facilitates desired outcomes, ensuring a focused and purposeful approach to AI interactions.

Before: Outlines the current situation or problem as the starting point, whether it's a problem, a desired skill, a task, or an aspiration for improvement. It provides context for the AI to understand their objectives.

After: Signifies the ideal solution or improvement from the "Before" situation, defining the desired outcome, goal, or state aimed to be achieved, such as mastering calculus concepts or maintaining productivity while working from home.

Bridge: Directs AI on specific actions to transition from the "Before" to the "After" state, connecting needs with desired outcomes. For instance, it might instruct AI to generate content, provide advice, or create a plan tailored to the goal.

By systematically outlining the current situation, defining desired outcomes, and providing actionable directives, this framework ensures that AI prompts are clear, purposeful, and tailored to address specific needs or challenges.

Context: Provides the background information or situation for AI model. It sets the stage for understanding the task to be performed or the desired output. By furnishing necessary details, the context ensures alignment between the prompt and the task at hand.

Action: Specifies the desired behavior or task to be performed, outlining the action that AI models need to take to achieve the desired outcome. It serves as a directive, guiding individuals on what steps to follow or what decisions to make in order to fulfill the objective.

Result: Specifies the desired outcome and provides a clear benchmark for success. For instance, in sentiment analysis, the result may be to classify text as positive or negative. It guides AI models towards achieving this outcome, ensuring clarity and alignment with the task.

Example: Provides concrete illustrations of prompt formulation and execution. For example, in text generation, it may offer a sample input sentence and its corresponding desired output. This demonstrates how the context, action, and result align to form a coherent prompt, aiding AI models in understanding and applying prompts effectively.

Role: Specifies the persona or position that the AI should adopt when generating responses. For instance, if the role is defined as a customer service representative, the AI's responses would be tailored to providing assistance and resolving inquiries. This ensures responses match the role, improving relevance and targeting based on context.

Input: Refers to the initial context or information provided to the model to guide its generation. It includes any relevant details, cues, or instructions necessary for the model to understand its task.

Steps: These are the individual steps or actions the model is expected to take in response to the input provided. Each step represents a sequence of interactions or decisions made by the model during the generation process.

Expectation: Outlines the desired outcome or output expected from the model after it has gone through the specified steps. It could be a response, a completion, a recommendation, etc.