Skip to content

Claude Skills by sumleo

github.com/sumleo

2 skillsA× 20 installs0 views

DocsA> Drop-in skill for Claude Code, Codex, and other agent environments. The TypeSafe agent skill gives your AI coding agent full context on the TypeSafe API: the three question [types](primitives.md), the architectural [patterns](patterns.md), and best practices for structuring evaluations.ai-agentsgobashVotes: 0GitHub stars: 3
Prompt2jevAUse when a user asks to convert an LLM prompt, system prompt, prompt template, or prompt-and-parse step (classifier, router, judge, grader, extractor, guardrail) into TypeSafe Jev decisions, points at the code that runs one (an LLM call that returns labels, scores, booleans, or JSON fields, with its parser and the branches on its output) and wants it replaced with typed questions, describes a decision requirement in natural language with no prompt yet, or asks for a Python script, TypeScript ...ai-agentsjavascripttypescriptVotes: 0GitHub stars: 3

Claude Skills by sumleo — 2 skills, security-graded | Skills Directory