How does the --research flag work in Lyra?

When the research flag is used, Lyra performs a web search to find current best practices for the specific type of prompt you are creating, ensuring the optimization uses the most up-to-date techniques.

What is the Lyra skill for Claude Code?

Lyra is an advanced prompt engineering assistant that helps you transform basic requests into high-performance prompts tailored for specific AI models like Claude, ChatGPT, and Gemini.

Does Lyra support platform-specific prompt techniques?

Yes, Lyra can optimize prompts specifically for Claude using XML tags and reasoning requests, or for ChatGPT using system messages and structured output constraints.

What is the difference between BASIC and DETAIL mode?

BASIC mode provides a fast, one-shot optimization for simple tasks, while DETAIL mode uses an interactive process to clarify your audience, constraints, and goals before generating the final prompt.

Can I use Lyra to improve my existing prompts?

Absolutely. You can provide an existing prompt to Lyra and ask it to 'optimize' or 'improve' it, and it will apply the 4-D Methodology to enhance its clarity and effectiveness.

Lyra AI Prompt Optimizer

Name: Lyra AI Prompt Optimizer
Author: joaquimscosta

byjoaquimscosta

•

Data Science & ML

Transforms vague natural language inputs into precision-engineered prompts for Claude, ChatGPT, Gemini, and other large language models.

Lyra is a specialized prompt engineering skill designed to bridge the gap between human intent and AI performance. By applying a systematic 4-D Methodology—Deconstruct, Diagnose, Develop, and Deliver—Lyra analyzes user requirements, identifies clarity gaps, and constructs highly structured prompts tailored for specific platforms. Whether you need a quick optimization via BASIC mode or a deep-dive interactive session using DETAIL mode, Lyra ensures your AI interactions are efficient, context-rich, and optimized for high-quality outputs, incorporating platform-specific techniques like XML tagging for Claude or system message structures for ChatGPT.

Key Features

014-D Methodology (Deconstruct, Diagnose, Develop, Deliver) for systematic prompt construction.

02Platform-specific optimizations for Claude, ChatGPT, and Gemini to leverage unique model strengths.

03Interactive DETAIL mode that asks targeted questions about audience, constraints, and outcomes.

04Optional web research integration to incorporate real-time prompt engineering best practices.

05Transparent reporting that explains what was improved and why, including specific techniques used.

063 GitHub stars

Use Cases

01Refining prompts specifically for Claude to utilize XML tags and long-context capabilities.

02Converting simple one-sentence requests into comprehensive technical prompts for complex tasks.

03Researching and applying the latest 2025 prompt engineering patterns for specialized domain-specific tasks.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add joaquimscosta/arkhe-claude-plugins lyra

For use in Claude.ai and ChatGPT

Download Skill