Skip to content
Design
Command

/design-prompt

Create a structured system prompt for an AI feature.

BOOST
From plugin
ai-design-skills
18818 skills18 commands
Install
> /plugin marketplace add owl-listener/ai-design-skills

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/design-prompt

Context preview

What this command does when you run it.

Create a structured system prompt for an AI feature.

Command definition

design-prompt.md
description: Create a structured system prompt for an AI feature.
argument-hint: "[AI feature or product to design a system prompt for]"

You are designing a system prompt. Use only skills from the prompt-architecture plugin. Follow this process:

Step 1: Define Requirements

  • What is this AI feature for?
  • Who are the users?
  • What should the AI do? What should it NOT do?
  • What output format and quality is expected?

Step 2: Structure the Prompt

Using **system-prompt-structure**:

  • Write the Identity and Role section
  • Write the Context and Knowledge section
  • Write the Behavioral Rules section
  • Write the Output Specifications section
  • Ensure sections are clearly separated and ordered by importance

Step 3: Design Constraints

Using **constraint-specification**:

  • Define format constraints (output structure, required fields)
  • Define length constraints (word count ranges, section proportions)
  • Define content constraints (topics to include/exclude, source restrictions)
  • Define tone constraints (formality, voice, audience)
  • Define quality constraints (accuracy, completeness, actionability)
  • Establish constraint priority hierarchy

Step 4: Add Examples

Using **few-shot-patterns**:

  • Create 2-3 input-output examples demonstrating ideal behavior
  • Include one common case and one edge case
  • Ensure examples are high quality and consistent with the constraints

Step 5: Design Context Integration

Using **context-engineering**:

  • Define where retrieved context will be injected
  • Specify context selection criteria
  • Allocate the context budget across sections
  • Design the information architecture within the prompt

Step 6: Plan for Iteration

Using **prompt-versioning**:

  • Document the rationale for key design decisions
  • Define test cases for evaluating the prompt
  • Establish the review and deployment process

Output

Deliver a complete system prompt package: 1. The system prompt itself (ready to use) 2. Prompt design rationale document 3. Constraint specification 4. Example library (2-3 examples) 5. Context integration specification 6. Test cases for validation 7. Version notes (v1.0)

Read more
Ships withai-design-skills

Part of the Designer Skills suite. Install this collection and four more with one command in Claude Code: /plugin marketplace add Owl-Listener/designer-skills This repo still works on its own. The suite just gives everyone one front door.

Get the whole plugin

Other commands on ai-design-skills.