Claude Prompt for Structured Output & Tool Schemas
Design JSON Schema (Draft 2020-12), LLM prompt, and validator for extracting KYC identity fields from PDFs with tables with strict schema adherence.
More prompts for Structured Output & Tool Schemas.
Design Pydantic BaseModel, LLM prompt, and validator for extracting job postings (title, salary, location, skills) from HTML docs with strict schema adherence.
Design JSON Schema (Draft 2020-12), LLM prompt, and validator for extracting bug reports from support chats from Confluence wikis with strict schema adherence.
Robust retry/repair loop that recovers from null vs empty string inconsistency in LLM JSON Schema (Draft 2020-12) output without looping or masking bugs.
Design Pydantic BaseModel, LLM prompt, and validator for extracting purchase orders from Jira tickets with strict schema adherence.
Design Pydantic BaseModel, LLM prompt, and validator for extracting purchase orders from research papers with strict schema adherence.
Design Pydantic BaseModel, LLM prompt, and validator for extracting insurance claim line items from regulatory filings with strict schema adherence.
Replace the bracketed placeholders with your own context before running the prompt:
[// 8-15 fields specific to KYC identity fields]— fill in your specific // 8-15 fields specific to kyc identity fields.[// if applicable — array of sub-records]— fill in your specific // if applicable — array of sub-records.[str]— fill in your specific str.