v0.3.3 - CLEAR Clone Implementation

CLEAR Clone Schema & API Implementation

This release implements comprehensive CLEAR-style reporting capabilities to compete with Thomson Reuters CLEAR at a fraction of the cost.

New Database Schema (ddl/05_clear_clone.sql)

Person Tables:

• person_alias - Track AKAs, maiden names, nicknames
• person_address_link - Historical address tracking with sources
• person_flags (materialized view) - YES/NO quick analysis flags:
  • OFAC listing
  • Criminal record
  • Bankruptcy
  • Sex offender
  • Multiple SSN
  • Deceased
  • P.O. Box address
  • Prison address
  • Younger than SSN issue
• person_person_rel - Person-to-person relationships (relative, associate, shared_address, shared_business)

Business Tables:

• business_phone - Business phone numbers with sources
• business_address_link - Business address history

New API Endpoints (api/clear_clone.py)

GET /clear/person/{person_canon_id}
Returns comprehensive person report with:

• Subject information (name, DOB, gender, entity ID)
• Aliases list
• Address history with flags
• Quick analysis flags (all boolean)
• Criminal records with case details
• Associate analytics

GET /clear/business/{business_canon_id}
Returns comprehensive business report with:

• Business subject info (name, DUNS, address)
• Business flags
• Phone numbers
• Officers list
• Same-address businesses
• Same-address people

Key Features

• JSON API returns every section found in $75 CLEAR PDF reports
• Complete audit trail with source citations
• Performance-optimized with indexed materialized views
• Ready for Redis caching layer (Week 4)
• Foundation for PDF export (Week 4)

Next Steps (Week 4 Backlog)

• Populate person_alias from person_raw during ingestion
• Populate person_address_link from address_raw nightly
• Seed business_phone & officer tables from SOS filings
• Redis cache for sub-5ms repeat lookups
• PDF generator using weasy-print
• Bulk CSV endpoint (10,000 lookups / 3s)