Audit Domain

Step 3: Analyze Relationship Model

Read family/relationship types in internal/domain/:

• Check relationship typing (biological, adopted, step, foster)
• Verify directional relationships are queryable
• Look for edge case handling (unknown parentage, complex families)

Step 4: Analyze Date Handling

Search for date-related types:

• Check for genealogical date types (exact, approximate, range, before/after, between)
• Check for partial date support (year-only, month-year)
• Look for calendar system considerations

Step 5: Analyze GEDCOM Handling

Read internal/gedcom/:

• Check import for data preservation (invariant DI-003 — lossless import)
• Check export for valid GEDCOM output
• Look for round-trip testing
• Check edge case handling (Unicode, non-standard tags, large files)

Step 6: Analyze Citation/Source Model

Search for source/citation types:

• Check if citations can attach to specific claims (not just people)
• Look for source/repository/citation distinction
• Evaluate against Evidence Explained patterns

Step 7: Check GPS Support

Evaluate whether the data model supports the 5 GPS elements:

1. Reasonably exhaustive search
2. Complete citations
3. Analysis and correlation
4. Resolution of conflicts
5. Soundly written conclusion

Output Format

Domain / Genealogy Audit Report

Scorecard

Top Findings

List up to 10 findings, risk-ranked. For each:

• Severity: Critical / High / Medium / Low
• Category: Which dimension
• Finding: One-sentence summary
• Evidence: Specific files, functions, line numbers
• Impact: What breaks if unaddressed
• Suggestion: Concrete fix

Issue-Ready Tickets

Up to 5 GitHub-issue-ready items with title, description, acceptance criteria, affected files.

Manual Verification

Up to 5 items requiring human judgment.