Question 206
Scenario: Structured Data Extraction
A system extracts structured fields from messy documents, validates every output against JSON schemas, handles edge cases gracefully, and stays accurate enough to feed downstream systems.
Read the full scenario lessonA developer is configuring tool_choice for a document extraction pipeline. The first step must always call the extract_metadata tool, but subsequent steps should let Claude decide which tools to use based on document content. Which tool_choice configuration sequence is correct?