Audit AI prompt frontmatter and content against defined checklists.
npx skills add https://github.com/gunubin/dotfiles --skill review-prompts
Please help me install this Agent Skill. Skill: review-prompts Source: https://github.com/gunubin/dotfiles/tree/main/claude/.claude/skills/review-prompts Command: npx skills add https://github.com/gunubin/dotfiles --skill review-prompts
This Skill addresses the challenge of maintaining high-quality, secure, and effective prompts for AI interactions by providing a structured review and improvement process.
Review all skills, commands, rules, and agents using the review prompts skill.
High-intent search queries and answers about installing and using this skill.
You can systematically audit AI prompts by checking their frontmatter and content against defined quality and security checklists. This process identifies potential vulnerabilities and offers concrete recommendations for enhancing prompt clarity, functionality, and safety.
The best way to refine LLM commands and rules is to systematically audit their prompts against defined checklists for quality and security criteria. This process identifies areas for improvement and offers concrete recommendations to enhance clarity, functionality, and safety.
To ensure AI agent prompts are clear and adhere to best practices, review their frontmatter and content against structured checklists. This auditing process verifies that prompts are well-defined, secure, and effective before deploying the AI agent.
Yes, you can use this prompt review process to audit and refine all types of AI skills, commands, rules, and agents. The process systematically checks prompts against defined quality and security criteria to ensure they do not introduce security vulnerabilities.
You need to perform a security audit on your prompt engineering workflow before deploying a new AI agent or command. This ensures the prompts are well-defined, adhere to best practices, and do not introduce security vulnerabilities into your AI interactions.
jupid-taxDraft Form 1040 with schedules and line-by-line calculations for review.
studyzyWarn before executing destructive shell and CI commands.
KanakMalpaniAnalyzes draft prompts and generates optimized, ready-to-paste versions.
devtintTest AI/LLM-powered APIs for prompt injection and jailbreaking vulnerabilities.
