instructor - Extract and validate structured LLM outputs
Extracts typed data from LLM responses, validates it with Pydantic, retries failures, and streams partial results.
Tags
Updated: 2026-10-02Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Extract structured data
- Validate Pydantic models
- Retry failed extractions
- Parse typed JSON
- Stream partial results
- Support multiple providers
Inputs
- Source text or prompts
- Pydantic response schemas
- Provider and model configuration
- API credentials
- Ollama endpoint
Outputs
- Validated structured data
- Validation error feedback
- Retry requests
- Partial streamed objects
- Streamed iterable items
Requirements
- Python environment
- instructor package
- pydantic package
- OpenAI or Anthropic package
- Supported LLM provider access
- Ollama endpoint for local models