What problem does it solve? LLM responses arrive as unstructured text that breaks downstream code when parsed manually. This Skill wraps LLM API calls so outputs are validated against Pydantic schemas, automatically retried with error feedback when validation fails, and streamed as partial typed objects. ## Core Features & Use Cases - Schema-Validated Extraction: Define Pydantic response models with field constraints, enums, nested models, and custom validators so LLM outputs always match your expected structure. - Automatic Retry with Feedback: When validation fails, the error message is sent back to the LLM and generation retries up to a configurable limit. - Streaming Partial Results: Stream partial objects or iterables as the LLM 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 name, founding year, industry, and employee count from hundreds of unstructured text snippets into typed Python objects ready for database insertion. ## Quick Start Use the instructor skill to extract a person's name, age, and email from this text into a validated Pydantic model.
