Create and evaluate Claude Code skills with SKILL.md files.
npx skills add https://github.com/bwgz/teambuilder --skill skill-creator-bwgz
Please help me install this Agent Skill. Skill: skill-creator Source: https://github.com/bwgz/teambuilder/tree/main/teambuilder-sdd/.claude/skills/skill-creator Command: npx skills add https://github.com/bwgz/teambuilder --skill skill-creator-bwgz
Automates the end-to-end process of creating new skills, editing existing ones, and evaluating their performance, enabling teams to iterate quickly from draft to deployed skill with measurable triggering accuracy.
Draft a new skill by outlining its purpose, create a SKILL.md with a clear name and description, and begin an eval loop to iteratively improve it.
High-intent search queries and answers about installing and using this skill.
You can create and refine Claude Code skills by capturing intent, writing a structured SKILL.md, running eval loops, and optimizing the description to improve triggering accuracy. This structured workflow ensures measurable quality from draft to deployment.
Evaluating prompt engineering accuracy involves creating test prompts, running eval loops, and analyzing the results to iteratively rewrite the skill. Description optimization based on these eval results maximizes the triggering accuracy of the skill.
An iterative eval workflow improves skill creation by running test prompts, analyzing failures, and rewriting the SKILL.md. This loop refines the skill's intent capture and optimizes its triggering description for accurate activation.
Yes, pyyaml is a required dependency for this skill-creation workflow. You need the pyyaml package installed in your environment to parse configurations and generate the structured SKILL.md files during the intent capture phase.
Yes, you can edit existing Claude skills by resuming the structured workflow. The process supports modifying existing SKILL.md files, running new eval loops, and optimizing the description to improve triggering accuracy for updated requirements.
AzrinPutraGuide exploratory conversations to elicit project requirements and generate an intent brief.
lucasfthDesign, validate, and iteratively improve Claude Skills with SKILL.md and eval loops.
XCmiaowDraft SKILL.md files and benchmark trigger accuracy with eval runs.
yasuwrldhyperCreate and refine AI skills with SKILL.md metadata and benchmark testing.
