feat(0126-HipGluteGroinLandmarkSpecsCaptureCompare): hip/glute capture, compare, placement

[Ryan-AI-Studios]

What

Track 0126 hip/glute/groin form-read capture/compare/adjust.

• Frozen v1 ids (ASIS/PSIS/glute_outer/bottom/top_seam/groin_fold/thigh_medial) in KNOWN_LANDMARK_IDS + template blanks
• Fuse: same-id left Y overlay + back X/Z (no new DEPTH_PAIRS)
• Sidecar hip_glute_metrics.json (proportion report stay 1.2.0)
• CLI proportion blockout-hip-glute-compare + MCP mesh_proportion_blockout_hip_glute_compare (catalog 50)
• Soft new-id glute Y/Z after seat write and dual lock (B33); do not abs measured (B32)
• Hold 0106/0092/0068/0053/0070/0036; no glute_outer X consume; no remake/D7

Honesty

HIP_GLUTE_COMPARE_HONESTY = authoring QA only. Green compare is not mesh or print success.

Why

Package A has hip/glute DEPTH + HAVE ids, not ASIS/PSIS/outer/bottom/groin form-read. Agents were guessing seat/span/gap from H-frac priors (0106/0092/0068).

Tests

ledgerful verify: ruff / format --check / basedpyright / pytest 2119 passed, 3 skipped. Codex R2 PASS.

---

Note

Medium Risk
Touches landmark fuse and glute recipe placement, which can change blockout geometry when new ids are present. Compare is authoring-only JSON and does not alter auth or print pipelines.

Overview
Adds track 0126 hip/glute/groin form-read: frozen landmark ids, fuse XYZ, authoring compare, and measured glute Y/Z overlay after seat dual-lock.

New ids (asis_*, psis_*, glute_outer/bottom, glute_top_seam, groin_fold_*, thigh_medial_*) land in known ids and template blanks. Fuse fills back X/Z and left Y without new DEPTH_PAIRS and without inventing missing Y.

blockout-hip-glute-compare (CLI + MCP; catalog 50) writes hip_glute_compare.json plus sidecar hip_glute_metrics.json. Compare is authoring QA only (HIP_GLUTE_COMPARE_HONESTY); report schema stays 1.2.0. Soft-adjust is limited to new-id glute Y/Z; outer X and existing hip/glute priors stay hold.

Recipe emit overlays measured glute_bottom_* / glute_top_seam Y/Z after dual lock so seat write cannot clobber them. Existing hip/glute scale constants are unchanged.

^{Reviewed by Cursor Bugbot for commit 4d98ef1. Bugbot is set up for automated code reviews on this repo. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 2 potential issues.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 4d98ef1. Configure here.}

[cursor]

Midline gap always flags

Medium Severity

glute_midline_gap treats any positive inner-surface separation as a problem. Normal dual RECIPE_glute_soft_* parts leave an intentional cleft, so product-class recipes always trip this token, unlike sibling form-reads that use real thresholds.

Additional Locations (1)

• src/meshops/proportion/hip_glute_compare.py#L633-L635

 

^{Reviewed by Cursor Bugbot for commit 4d98ef1. Configure here.}

[cursor]

Seam soft-adjust uses wrong X

Medium Severity

glute_top_seam is soft-adjust eligible and mapped to RECIPE_glute_soft_l. Back-fused seam X near midline versus the left glute center yields a large delta.x, so _suggest returns soft_adjust even when Y/Z already match. Soft consume only writes Y/Z.

Additional Locations (2)

• src/meshops/proportion/hip_glute_compare.py#L74-L75
• src/meshops/proportion/hip_glute_compare.py#L92-L98

 

^{Reviewed by Cursor Bugbot for commit 4d98ef1. Configure here.}