Prompt optimizer

Skill learn-with-santosh/claude-master-skills/skills/prompt-optimizer

A curated collection of specialized skills and workflows designed to enhance Claude's capabilities in specific domains. These skills provide frameworks, psychological triggers, and structured processes to deliver high-quality, professional results.

Install

npx -y skills add learn-with-santosh/claude-master-skills --skill prompt-optimizer

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Acts as a Prompt Refinement Engine. Use this skill whenever a user says "optimize this prompt", "make this prompt better", "I'm not getting good results from this prompt", or wants to upgrade a basic instruction into a professional-grade prompt.

SKILL.md

2.7 KB, as published. Nobody here has run it

Prompt Optimizer 💎

Turn your basic instructions into high-performance prompts. This skill uses advanced engineering techniques (Chain-of-Thought, Delimiters, Few-Shot, and Persona) to ensure your AI interactions are consistent, accurate, and powerful.

When to use this skill

Use this skill when:

You have a simple prompt that is producing vague or incorrect results.
You want to turn a task description into a reusable system prompt.
You need to add constraints or structural requirements to an AI output.
You want to "AI-ify" a business process.

The Optimization Framework

To optimize a prompt, Claude applies these layers:

1. The Persona Layer

Define who Claude should be. "Instead of 'Write a script', use 'You are a Senior Python Developer with 10 years of experience in automation...'"

2. The Objective Layer

Clarify exactly what the goal is. "Define the 'Primary Mission' and the 'Success Criteria' for the task."

3. The Constraint Layer

Add the "Boundaries."

Tone (Professional, Humorous, Academic)
Length (Max 200 words, exactly 5 bullet points)
Format (JSON, Markdown, CSV)

4. The Engineering Layer

Add advanced logic:

Chain-of-Thought: Add <thought> blocks.
Delimiters: Use tags like <context> or <instructions>.
Few-Shot: Request specific examples to be added.

Output Format

🛠 Analysis

What was wrong with the original prompt?

✨ The Optimized Prompt

The ready-to-use, engineered version.

💡 Why it's better

Explanation of the techniques applied.

Examples

Example: Transforming a simple request

Original: "Help me write a blog post about AI." Optimized: "Persona: You are a Tech Journalist and SEO Expert. Objective: Write a 1,000-word deep-dive article about the impact of Claude Code on developer productivity. Instructions:

Lead with a 'Pattern Interrupt' hook.
Use H2 headers for each major section.
Include a 'Key Takeaways' summary at the end. Constraints: Avoid buzzwords. Use a conversational but professional tone."

Best Practices

Iterate: If the first optimization isn't perfect, tell Claude what's missing.
Use Prompt Pro: For extremely complex prompts, use the prompt-engineer-pro skill first to define the strategy, then prompt-optimizer to draft it.
Test: Always run the optimized prompt through its paces with different inputs.