Create, modify, and evaluate AI skills with Claude tools.
npx skills add https://github.com/eastreams/loong --skill skill-creator-eastreams
Please help me install this Agent Skill. Skill: skill-creator Source: https://github.com/eastreams/loong/tree/main/skills/skill-creator Command: npx skills add https://github.com/eastreams/loong --skill skill-creator-eastreams
It streamlines the end‑to‑end process of creating, iterating, and benchmarking AI skills, removing the manual overhead of drafting, testing, and optimizing skill descriptions.
Use the skill‑creator to draft a new skill for extracting data from PDFs and run an evaluation.
High-intent search queries and answers about installing and using this skill.
Create and test AI skills by drafting SKILL.md files, generating test prompts, running evaluations, and grading results to build custom vertical agents. This streamlines the end-to-end process from initial intent capture to iterative refinement.
AI skill benchmarking compares skill-enabled runs with baselines to produce benchmark reports, improving description triggering accuracy. It systematically evaluates whether your skill activates correctly for intended queries and optimizes its performance over time.
Optimize AI agent skill descriptions by using benchmark reports to compare skill-enabled runs against baselines, then iteratively refining the SKILL.md instructions. This description optimization process directly improves triggering accuracy and evaluation metrics.
Yes, this supports rapid prototyping of new AI skills with Claude tools by guiding you to capture intent, write SKILL.md files, and run quick evaluations. It is designed for developers, product managers, and AI engineers needing fast iteration cycles.
The best way to evaluate custom AI agent skills is to generate test prompts, run evaluations, collect metrics, and grade results systematically. This skill handles the end-to-end evaluation process, removing manual overhead and enabling iterative improvement.
No prior benchmarking experience is required, but developers, product managers, or AI engineers will find it most suitable. The skill guides you through drafting, testing, and benchmarking, systematically removing the manual overhead of optimizing skill descriptions.
huangservaRoute user intent to domain-specific Skills and generate structured prompts.
gclmGuide Claude Code skill development through SKILL.md creation and test cases.
gesmith0606Create, evaluate, and iteratively improve SKILL.md descriptions with Python scripts.
HYP3R00TOrchestrate end-to-end creation, evaluation, and improvement of Claude skills.
