highlight audited YOLO label relationships

docs/CODEX_EXECUTION_LOG.md

@@ -12565,3 +12565,33 @@ Open:
   on a dedicated 384 px contact sheet. Visual follow-up found complex/adjacent
   building components rather than a systematic duplicate-label pattern.
 - Targeted auditor and renderer tests passed: 9 tests.
+
+## 2026-08-09 - Index-bound relationship highlighting
+
+### Improved
+
+- Connected the relationship-audit manifest to the visual label renderer so
+  exact duplicates, near duplicates and possible nesting receive distinct
+  colors and every implicated YOLO row is labelled with its source index.
+- Generated a new immutable 36-tile focused sheet on Tower. It renders 17,253
+  valid labels, highlights all 86 rows participating in the 43 nesting pairs
+  and contains zero invalid label rows.
+- AI-assisted inspection confirmed that the focused corpus contains only
+  magenta nesting candidates; no exact or high-IoU duplicate category is
+  present. The cases remain dispersed dense-GRB review candidates rather than
+  a safe deterministic rewrite set.
+- Retained every source file and label unchanged. Excluding entire flagged
+  tiles would remove 17,167 unflagged labels and is therefore not justified.
+
+### Verified
+
+- Targeted relationship-auditor, renderer and Sprint 167 regression suite:
+  10 tests passed.
+- Scoped Ruff check, Ruff format check, Python compilation and Git whitespace
+  check passed.
+
+### Remaining limitation
+
+- This is explicit AI-assisted triage, not human adjudication. It improves the
+  evidence and review precision but does not unlock governed production
+  training or checkpoint promotion.

docs/TODO.md

@@ -1157,6 +1157,8 @@ This file now starts with the current implementation status. Older preparation/b
   legitimately.
 - [x] Persist a deterministic label-relationship audit and a focused 36-tile
   contact sheet; AI follow-up found no systematic duplicate-label pattern.
+- [x] Bind the relationship audit back to exact label-row indices and render
+  all 86 implicated rows with type-specific highlighting for focused review.
 - [ ] Convert the AI-assisted ledger into no stronger claim than experimental
   triage; a real human must independently review and sign the frozen artifacts
   before the governed training wrapper may unlock.

docs/reviews/2026-08-09-ai-assisted-checkpoint-review.md

@@ -117,3 +117,15 @@ AI-assisted inspection found no systematic duplicate-label pattern: the
 relationships predominantly represent adjacent or complex building components
 in dense GRB contexts. All tiles remain available for experimental analysis,
 while the exact 43 pairs stay visible for human release adjudication.
+
+A second focused render now binds the relationship manifest back to the exact
+YOLO label-row indices. It highlights all 86 implicated rows: cyan for exact
+duplicates, red for near duplicates and magenta for possible nesting. The sheet
+contains only magenta relationship marks, confirming visually and
+machine-readably that there are no exact or high-IoU duplicates in this set.
+The nesting cases are geographically dispersed and commonly combine a larger
+GRB object envelope with a smaller component. They are not deterministic
+rewrite candidates. Excluding the 36 complete tiles would also discard 17,167
+unflagged labels, so no automatic tile exclusion or label mutation was made.
+The checksum-bound highlighted sheet and summary are recorded in the AI review
+ledger.