Create, edit, and optimize AI agent skills with evaluation tools.
npx skills add https://github.com/Luohao-Yan/warp-cn --skill create-skill-luohao-yan
Please help me install this Agent Skill. Skill: create-skill Source: https://github.com/Luohao-Yan/warp-cn/tree/main/resources/bundled/skills/create-skill Command: npx skills add https://github.com/Luohao-Yan/warp-cn --skill create-skill-luohao-yan
This Skill unit assists users in creating, editing, and optimizing skills, thereby improving the functionality and efficiency of AI agents.
To create a new skill for code review, use the create-skill unit by following the provided guide and running the necessary commands to define your skill's behavior and evaluate its effectiveness.
High-intent search queries and answers about installing and using this skill.
To create and optimize AI agent skills for code review, you draft the skill, test it with sample prompts, and iterate based on performance feedback using dedicated measurement tools.
Iterative improvement of skill performance involves running evaluation scripts to benchmark current behavior, identifying weaknesses, and editing the skill definition to achieve better agent interaction.
Yes, skill editing allows you to modify and enhance existing AI agent skills directly, updating their behavior and definitions without needing to rebuild the entire skill from scratch.
Performance measurement works by utilizing provided scripts and tools to benchmark the skill against sample prompts, generating measurable feedback to guide the refinement of agent capabilities.
Yes, understanding skill development processes and AI interaction is required to effectively define behavior, run measurement scripts, and interpret performance feedback for iterative improvement.
merlyn-huntressProvide a basic sample skill structure for demonstration and templating.
timurgaleevMonitor post-deploy web apps for defects and performance regressions.
nicolanoviellCreate, modify, and benchmark AI skills with Python scripts.
liuerfireDesigns, optimizes and evaluates prompts for LLMs using techniques like chain-of-thought and few-shot learning.
