Instructor

This wraps the Instructor library, which forces LLMs to return properly structured, validated outputs instead of raw text you have to parse yourself. It's built on Pydantic schemas, so you define what you want once and it handles extraction, validation, and retries automatically. Works across different LLM providers with the same API. The 15,000 GitHub stars and 100,000+ developer claim suggest it's legitimately popular, though those numbers are from the upstream library, not this specific skill wrapper. If you're tired of writing brittle JSON parsing code or dealing with hallucinated field names, this is worth trying. Best for data extraction tasks where output consistency actually matters.

Install to Claude Code

npx -y skills add davila7/claude-code-templates --skill instructor --agent claude-code