review-prompt

Name: review-prompt
Availability: InStock
Author: mikanfactory

Analyze LLM prompts for quality, efficiency, and safety issues.

Updated Jan 6, 2026

Search Tags:#code review #prompt engineering #prompt optimization #llm #evaluation #ai safety

Ready to run

npx skills add https://github.com/mikanfactory/dotfiles --skill review-prompt-mikanfactory

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: review-prompt
Source: https://github.com/mikanfactory/dotfiles/tree/main/home/dot_claude/skills/review-prompt
Command: npx skills add https://github.com/mikanfactory/dotfiles --skill review-prompt-mikanfactory

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates the comprehensive review of LLM prompts, ensuring they are clear, efficient, safe, and effective, thereby improving AI output quality and reducing risks.

Core Features & Use Cases

Automated Prompt Analysis: Evaluates prompts across clarity, efficiency, safety, pattern effectiveness, and best practices.
Risk Identification: Detects vulnerabilities like prompt injection and data exfiltration.
Use Case: Before deploying a new AI agent, use this Skill to rigorously test its system prompt for any potential security flaws or areas where its instructions could be misinterpreted, leading to better performance and safety.

Quick Start

Review the staged prompt files for quality and safety concerns.

Frequently Asked Questions about review-prompt

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

FAQPage Schema

How do I evaluate LLM prompts for quality and safety before deployment?▼

Evaluating LLM prompts for quality and safety requires analyzing them across clarity, efficiency, and vulnerability to prompt injection. This Skill uses a structured framework to identify security flaws and provide actionable recommendations.

What is the best way to detect prompt injection vulnerabilities in AI agents?▼

Detecting prompt injection vulnerabilities involves scanning the prompt for potential data exfiltration risks and unsafe patterns. This Skill categorizes identified security flaws by severity, ensuring AI agents resist malicious inputs.

How do I analyze inefficient token usage in LLM system prompts?▼

Analyzing inefficient token usage requires checking prompts for unclear instructions and redundant patterns. This Skill evaluates prompt efficiency, highlighting areas to optimize token consumption without losing output quality.

Can I automate prompt optimization for better AI output quality?▼

Automating prompt optimization involves running a multi-category assessment on staged prompt files. This Skill evaluates pattern effectiveness and best practices, providing actionable recommendations to improve AI output quality.

Why does my AI agent misinterpret instructions and how can I fix it?▼

AI agents misinterpret instructions when prompts lack clarity or contain conflicting patterns. This Skill analyzes prompt files to identify unclear instructions, categorizing findings by severity to guide precise corrections.

What are the limitations of automated prompt evaluation for AI safety?▼

Automated prompt evaluation assesses staged prompt files for known vulnerabilities and inefficiencies, but cannot predict all runtime behaviors. Findings are categorized by severity to help prioritize manual review of edge cases.