prompt-optimizer

Audit and rewrite LLM prompts to improve clarity and alignment with user intent.

2|Updated Mar 26, 2026
One-click install
npx skills add https://github.com/calvingit/skills --skill prompt-optimizer-calvingit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prompt-optimizer
Source: https://github.com/calvingit/skills/tree/main/skills/prompt-optimizer
Command: npx skills add https://github.com/calvingit/skills --skill prompt-optimizer-calvingit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill audits and rewrites prompts for LLMs to improve clarity, reliability, and alignment with user intent across system prompts, user prompts, translation prompts, coding prompts, agentic system prompts, and any instruction given to an LLM.

Core Features & Use Cases

  • Audit prompts: identify ambiguities, safety risks, and prompt injection vulnerabilities.
  • Rewrite prompts: produce clear, structured, and effective prompts aligned with user goals.
  • Use cases: improving system prompts, user prompts, translation prompts, coding prompts, and any instruction given to an LLM.

Quick Start

Provide a prompt you want improved, and I will audit and rewrite it.

Frequently Asked Questions about prompt-optimizer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I fix an LLM prompt that is returning the wrong output format or ignoring instructions?

Auditing and rewriting the prompt improves its structure and clarity, directly fixing issues like wrong output formats, inconsistent results, partial execution, ignored instructions, and LLM hallucinations.

What is the best way to audit a system prompt for safety risks and ambiguities?

Auditing a system prompt identifies ambiguities, safety risks, and prompt injection vulnerabilities, ensuring the instruction is structured effectively to produce reliable LLM outputs across various use cases.

How do I rewrite an agentic system prompt to prevent partial execution and hallucinations?

Rewriting agentic system prompts involves restructuring the instructions to align tightly with user goals, eliminating ambiguities that cause partial execution, inconsistent results, and hallucinations.

Can I use prompt engineering techniques to improve translation and coding prompts?

Yes, prompt engineering techniques apply to translation and coding prompts by auditing them for clarity and rewriting the instructions to ensure reliable execution and accurate results from the LLM.

Why does my LLM produce inconsistent results from the same prompt?

LLMs produce inconsistent results from the same prompt due to structural ambiguities or misalignment with user intent, which can be resolved by auditing and rewriting the prompt for better clarity.