Build, test, and benchmark Claude Skills with evaluation workflows.
npx skills add https://github.com/juprem/sandbox-react --skill skill-creator-juprem
Please help me install this Agent Skill. Skill: skill-creator Source: https://github.com/juprem/sandbox-react/tree/main/.gemini/skills/skill-creator Command: npx skills add https://github.com/juprem/sandbox-react --skill skill-creator-juprem
Create, refine, and evaluate Claude Skills across their lifecycle: from ideation to benchmarking, iteration, and deployment.
Draft a new skill, then run the end-to-end eval loop to test and iterate until the triggering is robust.
High-intent search queries and answers about installing and using this skill.
To create and test Claude Skills, draft a new skill and run the end-to-end evaluation loop to measure triggering accuracy and iterate until deployment is reliable.
Skill triggering optimization is the process of running benchmarks and multi-step evaluation workflows to refine a skill's description, ensuring reliable and accurate activation across varied inputs.
Yes, you can benchmark multiple skill variants by running variant testing and automated optimizations, archiving each iteration with traceable results to compare performance.
Yes, pyyaml is a required dependency for running the skill creation, refinement, and automated evaluation workflows that handle configuration and traceable benchmarking results.
You integrate scripts and references by attaching them as optional resources during skill creation, supporting deterministic task execution and providing rich documentation for automated workflows.
samjhechtGenerate Constitution, Roadmap, and process documentation templates for project governance.
Tnemo65Analyze slow queries and execution plans to optimize database performance.
salverius-techCreates AI skills from defining intent to finalizing SKILL.md structure and tuning descriptions for triggering accuracy.
CeeKay1901Draft, test, and refine reusable skills with evaluation and trigger optimization.
