Plant Workspace Photo Consistency Validation

August 10, 2026

Added evidence-backed photo-reference and orphan detection to Plant Model validation.

Tags: Abbey Root

Plant Workspace Photo Consistency Validation

Summary

An AI backlog-leverage decision grouped eight plant tasks into one alleged session. The actual implementation boundary was smaller: canonical photo references, workspace photo inventory, and their set differences all belong to one read-only validation pass.

abbey plant validate now checks role paths, history photo entries, Markdown images, and photo-metadata tables against supported files in photos/. Missing references fail. Unreferenced files warn because preserved originals may be intentional source material rather than defects.

Accomplishments

• Added canonical reference discovery across facts and Markdown.
• Added missing-reference failures with source document and line evidence.
• Added warning-level undocumented/orphaned photo reporting.
• Ignored XMP, AppleDouble, and unsupported artifacts.
• Added focused regression fixtures and validated all eleven real workspaces.
• Found two genuine undocumented Doctor Robert photographs.

Lessons Learned

• Shared domain language is not enough to establish one implementation scope.
• Missing references and extra files have different risk semantics.
• Real-workspace validation is essential for tuning consistency rules and exposing genuine findings without inventing cleanup policy.

Next Steps

• Classify Doctor Robert’s two orphaned photographs.
• Address the separate missing sources/ directory debt for Phal McCartney and Something.