About shimo4228/when-code-when-llm
An Agent Skill (a folder with one Markdown instruction file that a coding agent loads when a task matches its description) for developers who mix plain code and LLM calls, and keep facing one recurring engineering question: for this one task, should I write deterministic code or call an LLM? It gives the agent a single decision axis (is the property being checked *structural* or *semantic*?), plus a false-positive test and worked examples of both failure directions.