What problem does it solve? Getting reliable, structured data out of LLM responses is error-prone: raw JSON parsing breaks on malformed output, and there is no built-in way to validate types or recover from bad generations. This Skill uses the Instructor library to bind LLM outputs to Pydantic models, automatically validating fields and retrying failed extractions with error feedback. ## Core Features & Use Cases - Schema-Bound Extraction: Define Pydantic response models with type hints, field constraints, enums, and nested models so LLM outputs are validated and type-safe. - Automatic Retry on Validation Failure: When output fails validation, the error is fed back to the LLM and the request retries up to a configurable limit. - Streaming Partial Results: Stream partial objects or iterable items as the model generates them for real-time UI updates. - Multi-Provider Support: Works with Anthropic Claude, OpenAI, and local models via Ollama using a consistent API. - Use Case: Extract company information (name, founded year, industry, headcount) from unstructured text into a validated CompanyInfo model, or classify support tickets into an enum of categories with a confidence score. ## Quick Start Ask the AI to extract a person's name, age, and email from a block of text into a validated Pydantic model using Instructor with your preferred LLM provider.
