Files
Chartwell/Books/Accounting/Starter.txt

92 lines
4.9 KiB
Plaintext
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
StarterQuestions Module RAG Knowledge Base
Chunk 1: Module Overview
Metadata: module=StarterQuestions, type=guidance, location=CMH.HFA.Accounting.Orchestration/src/StarterQuestions, purpose=questionInspiration
Purpose: Provides example questions and guidance for new users unfamiliar with the system. Helps users know what types of questions they can ask across modules.
Scope: Covers Quote, Repository, Services, and StructuredTags modules with illustrative questions, plus general guidance on asking questions.
Key Concepts: question examples, user guidance, RAG-friendly prompts, workflow understanding, module orientation.
Chunk 2: Quote Module Questions
Metadata: domain=Quote, type=exampleQuestions, purpose=quotePreprocessing
Sample Questions:
• How are insurance quotes converted to Oracle customer records?
• What Lambda function handles quote offered events?
• How is the primary bill-to site configured?
• How are customer addresses normalized for Oracle integration?
• What default values are applied during quote preprocessing?
• How is the AccountDescription field constructed?
• When does customer record creation occur in the quote workflow?
• What Oracle address set is used for HFA customers?
• Which fields from the quote payload map to Oracle customer records?
Chunk 3: Repository Module Questions
Metadata: domain=Repository, type=exampleQuestions, purpose=dynamoDBAccess
Sample Questions:
• How do I retrieve FIFO events by policy number?
• What is the repository pattern implementation for DynamoDB?
• How does conditional update work for FIFO event status?
• What table naming convention is used for repositories?
• How is TTL configured for completed events?
• How does rollback work when saving adjustment configurations?
• What concurrency control mechanisms prevent race conditions?
• How are adjustment configurations stored in DynamoDB?
• How do I query events by unique EventId?
Chunk 4: Services Module Questions
Metadata: domain=Services, type=exampleQuestions, purpose=adjustmentConfiguration
Sample Questions:
• How do I retrieve an Oracle adjustment configuration by name and type?
• How is active status evaluated for adjustment configurations?
• How do I save all adjustment configurations at once?
• What validation is performed during configuration retrieval?
• How do endorsement workflows use adjustment configurations?
• What is the difference between Retail, VMF, and Customer refund configurations?
• Which Lambda functions are available for configuration management?
• How does temporal validity work for adjustment configurations?
• What Oracle properties are stored in adjustment configurations?
Chunk 5: StructuredTags Module Questions
Metadata: domain=StructuredTags, type=exampleQuestions, purpose=glMapping
Sample Questions:
• How do I filter memo lines by commission percentage?
• How are structured tags parsed and evaluated?
• What dimensions can I use to filter GL account mappings?
• How does temporal validity work for memo line configurations?
• How do inclusions and exclusions (whitelist/blacklist) work?
• How do I calculate distribution amounts for transaction lines?
• Where do memo line configurations come from in the system?
• How are GL account mappings assigned for credit memos or invoices?
• How are Assurant vs non-Assurant rounding rules applied?
Chunk 6: General Guidance for Asking Questions
Metadata: domain=AllModules, type=userGuidance, purpose=questionTips
Guidance:
Think in terms of “What, How, When, Where”:
What data is processed?
How is it transformed?
When in the workflow does this occur?
Where is this information stored or sent?
Consider module boundaries to narrow your questions:
Quote → customer creation workflow
Repository → data persistence and retrieval
Services → Lambda endpoints and orchestration
StructuredTags → GL account mapping and configuration rules
Dont worry about phrasing perfectly. Partial matches or paraphrases often work.
Start with practical scenarios:
“I want to know how a customer record is created from a quote.”
“I need to find the configuration for a refund adjustment.”
“I need to see how memo lines are filtered for commission calculations.”
Chunk 7: Sample RAG Queries
Metadata: type=queryPatterns, purpose=RAGRetrieval, domain=AllModules
Sample Queries:
• “Show me how quote payloads become Oracle customer records.”
• “Retrieve FIFO events for a specific policy number.”
• “Explain the rollback mechanism for adjustment configurations.”
• “Filter memo lines by action=Issue and percent=70.”
• “What Lambda functions manage adjustment configurations?”
• “How is the bill-to site set up for new customers?”
• “What Oracle properties are stored for adjustment configurations?”
• “How does temporal validity affect memo line selection?”
• “How do inclusions and exclusions in structured tags work?”