Schema Evolution Impact Analysis Skill

When a model or database schema changes, trace the full downstream impact to identify all code that needs updating.

Trigger Conditions

• Model struct fields are added, removed, or modified
• New migration files are created
• Database schema configuration changes
• User invokes with "schema impact" or "schema-evolution-impact"

Input Contract

• Required: The changed model file and specific field changes
• Optional: Migration file for the change

Output Contract

• List of all files that reference the changed model/field
• Impact classification per file (breaking / needs-update / unaffected)
• Migration requirements (new migration needed, data backfill, etc.)
• Test files that need updating
• DTO/response files that need updating

Tool Permissions

• Read: All Go source files
• Write: None (read-only analysis)
• Search: Grep for model type names, field names, GORM tags

Execution Steps

1. Identify changes: Parse the diff to find added/removed/modified fields and their types
2. Trace references: Search for all files importing or referencing the changed model
3. Classify impact: For each reference, determine if it's breaking (uses removed field), needs-update (uses modified field), or unaffected
4. Check migration: Verify a migration exists for the schema change; check if data backfill is needed
5. Check DTOs: Verify request/response DTOs reflect the model changes
6. Check tests: Identify test files that create or assert on the changed fields
7. Report: Produce impact report with file:line references and required actions

Success Criteria

• All downstream impacts identified
• Migration exists for every schema change
• No orphaned references to removed fields

References

• .cursor/rules/122-gorm-conventions.mdc

• .cursor/skills/schema-migration/SKILL.md