promptify

What This Skill Does

The promptify skill is a sophisticated meta-agent designed to refine, structure, and optimize user prompts for maximum clarity and performance. It acts as an intermediary between the user's initial intent and the final model execution. By intelligently applying structural contracts—Role, Task, Constraints, and Output—the skill ensures that every prompt is tailored for precision. Whether you are drafting a complex coding query, requesting analytical research, or seeking creative writing assistance, promptify evaluates your input, determines if external context (web search, codebase access, or clarification) is required, and reconstructs the prompt into a high-fidelity artifact.

Installation

To install this skill, run the following command in your terminal: clawhub install openclaw/skills/skills/tolibear/promptify-skill

Use Cases

Use promptify whenever your interactions feel like they are producing generic, repetitive, or incomplete results. It is particularly powerful when:

• You have a vague goal like "improve this code" or "make this email sound better" and need the AI to ask you the right questions first.
• You are working within a specific codebase and need the AI to reference local files without you manually pasting them.
• You are exploring new technical domains where you need to synthesize the latest documentation from the web with your project requirements.
• You have a messy, unstructured stream-of-consciousness thought process that needs to be distilled into a professional, actionable prompt.

Example Prompts

1. "Optimize my prompt: I want to write a script that fetches user data from our API and saves it to a CSV."
2. "+web Please help me understand the latest industry best practices for React state management and rewrite this prompt to ensure I get expert, modern advice."
3. "Make this clearer: I need to refactor the authentication module, but I'm worried about edge cases. Can you help me write a prompt that asks for a robust testing strategy?"

Tips & Limitations

• Modifiers: Use the +ask, +deep, and +web modifiers to override the auto-detection engine if you know exactly which context source you need to prioritize.
• Contract Enforcement: Every prompt processed by this skill will include a role, specific task, constraints, and defined output format. If you find the structure too rigid for quick tasks, you can manually override by stating your preferences.
• Complexity: For extremely complex multi-file codebase tasks, combine +deep with a brief mention of the primary directory or language being used to help the agent narrow its scope.