ROLE: Skilled prompt engineer with a focus on cross-platform prompt design (text, picture, and multimodal AI systems).
TASK: Create a structured, reusable prompt template that, when applied to any AI system, consistently yields excellent results.
TEMPLATE STRUCTURE:
Include the following sections, each with a clear label:
1. PURPOSE
One or two phrases outlining the precise objective or result the prompt is intended to produce.
2. Context
Background data, such as domain, audience, circumstance, or use case, is necessary for the AI to accurately interpret the task. When details differ depending on the application, use placeholders (such as [industry] and [audience]).
3. PARAMETERS INPUT
Enumerate the variables (such as [subject], [format], [reference material], and [picture description]) that the user must provide in order to execute the template. Indicate whether each is necessary or not.
4. CONSTRAINTS AND LIMITATIONS
Length, scope, tone limitations, technical constraints (such as resolution for picture models and word count for text models), and platform-specific guidelines are all boundaries that the output must adhere to.
5. STYLE AND TONE
Describe the voice and stylistic register (such as formal, conversational, technical, or minimalist) as a placeholder that can be changed: [tone].
6. EXPECTED FORMAT OF OUTPUT
Indicate the precise format of the outcome according to the platform type (text/image/multimodal), including sections, bullet points, tables, image dimensions, JSON, etc.
DESIGN DEMANDS:
- The template should be brief, with no extraneous instructions or filler.
- Every section must have clearly specified placeholders for platform-specific customization and be platform-agnostic by default.
- Give precision and clarity precedence over length.
- Make sure the template can be applied to various projects by simply changing the placeholder variables.
GUIDE FOR CUSTOMIZATION:
Give a brief (three to five lines) set of instructions outlining how to modify the template for a new use case, including what to change for image, text, and multimodal systems, which placeholders to fill, and how to tighten or loosen constraints depending on task difficulty.
OUTPUT: Provide the customization guide after the completed template in a block that is ready to utilize.