Rename snaporca/SnapOrca to orca_cad so the OrcaSlicer PR carries no Snapmaker naming

2026-09-18 22:42:37 +00:00 · 2026-09-10 10:58:46 +02:00
parent 498e92ee35
commit 31a15cc5e5
44 changed files with 229 additions and 229 deletions
@@ -1,4 +1,4 @@
-# SnapOrca-CAD vs Onshape — capability gap analysis
+# Orca-CAD vs Onshape — capability gap analysis

 Generated 2026-07-22 by enumerating the source, not from recollection:
 `CadFeatureType` and `add_*` in `src/libslic3r/CAD/CadDocument.hpp`, `Tool` in
@@ -6,19 +6,19 @@ Generated 2026-07-22 by enumerating the source, not from recollection:
 `SketchConstraintType` + `SketchEntity::Type` in `src/libslic3r/CAD/SketchEngine.hpp`,
 and the JSON-RPC dispatch in `src/slic3r/GUI/CAD/McpControl.cpp`.

-**Scope note.** Onshape is a cloud PLM platform; SnapOrca is a Design tab inside a
+**Scope note.** Onshape is a cloud PLM platform; Orca is a Design tab inside a
 slicer. A large share of Onshape's surface (release management, branching, real-time
 collaboration, FEA, rendering, PDM) is out of scope by construction and is listed
 separately at the bottom rather than counted as a "missing tool".

 ---

-## 1. What SnapOrca already has
+## 1. What Orca already has

 ### 2D sketcher — near parity with Onshape
 This is the strongest area. Very little is missing.

-| Category | SnapOrca |
+| Category | Orca |
 |---|---|
 | Entities | Line, Polyline, Arc (3-point / tangent / center), Circle (center / 2-point / 3-point), Point, Ellipse, Elliptical arc, B-spline |
 | Shapes | Rectangle (corner / center / oblique / rounded), Slot, Arc-slot, Polygon |
@@ -88,7 +88,7 @@ Studio + configurations are a core differentiator, and this is the cheapest Tier
 item to close for the size of the payoff.

 **4. Surface modelling.** Absent. No surface extrude/revolve/loft/sweep, no fill,
-knit, trim/extend surface, offset surface, or thicken. SnapOrca is solid-only.
+knit, trim/extend surface, offset surface, or thicken. Orca is solid-only.
 *Impact:* organic/complex shapes and repair of imported junk geometry are impossible.
 OCCT already provides all of it (`TKOffset`, `TKBRep`), so the kernel is not the
 blocker — only UI and feature plumbing.
@@ -106,7 +106,7 @@ blocker — only UI and feature plumbing.
 | **Split body** | Cut removes material; splitting one body into two independently-usable bodies is absent. Very relevant for print-in-parts. | Medium |
 | **Thicken** | Solid from a surface/face offset. | Needs surfaces |
 | **Rib** | Standard structural feature. | Medium |
-| **Delete face / move face / replace face** | Direct/dumb-solid editing — the main tool for fixing imported STEP. Given SnapOrca imports STEP *and* meshes, its absence is felt. | Medium |
+| **Delete face / move face / replace face** | Direct/dumb-solid editing — the main tool for fixing imported STEP. Given Orca imports STEP *and* meshes, its absence is felt. | Medium |
 | **Datum axis, coordinate system** | Only datum *planes* exist. Axes are needed for revolve/pattern references. | Yes |
 | **Mass properties** | `GeometryEngine` computes a volume internally, but there is no volume/mass/COM/inertia readout. For print cost/time estimation this is nearly free to expose. | Yes — trivial |
 | **Measure tool in the GUI** | `measure` exists over MCP but there is no interactive measure in the UI. | Yes |
@@ -123,7 +123,7 @@ blocker — only UI and feature plumbing.
 Version control with branching/merging, release management, real-time multi-user
 collaboration, cloud PDM, FeatureScript custom-feature authoring, simulation/FEA,
 photorealistic rendering, app store/integrations. These are Onshape-the-platform,
-not Onshape-the-modeller. Not defects in SnapOrca.
+not Onshape-the-modeller. Not defects in Orca.

 ---

@@ -140,7 +140,7 @@ the most capability per unit of work:
 5. **Split body** — high value for print-in-parts workflows.
 6. **Project edges into sketch** — the sketcher's most conspicuous hole.
 7. **Surface modelling** — large, but OCCT already ships the algorithms.
-8. **Assemblies** — largest effort; only worth it if SnapOrca targets multi-part products.
+8. **Assemblies** — largest effort; only worth it if Orca targets multi-part products.

 Deliberately last: drawings and sheet metal — high cost, low relevance to an
 FDM-oriented tool.
@@ -149,7 +149,7 @@ FDM-oriented tool.

 ## 4. Honest summary

-SnapOrca's **sketcher is at or near Onshape parity**, and its **solid feature set
+Orca's **sketcher is at or near Onshape parity**, and its **solid feature set
 covers the mainstream modelling path** (sketch → extrude/revolve/sweep/loft →
 dress-up → boolean/pattern). What is absent is *breadth*: assemblies, surfaces,
 sheet metal, drawings, and — most importantly for a tool calling itself parametric —
@@ -4,7 +4,7 @@ What the Design tab actually costs a maintainer who merges it. Written to be che
 every number below is reproducible with the command that produced it, and the places where
 a number is still missing say so instead of guessing.

-Measured on Linux x86_64, OCCT V7_6_0, in the `snaporca-deps` build image.
+Measured on Linux x86_64, OCCT V7_6_0, in the `snapmaker-deps` build image.

 ## Summary

@@ -75,7 +75,7 @@ producing them honestly needs a build this machine cannot do:

 1. **Windows DLL delta.** OCCT builds shared on Windows, so the shipped cost there is real
   DLL bytes rather than linker-selected objects. That needs a Windows build to size —
-   tracked as the cross-platform build proof (`snaporca-gix`).
+   tracked as the cross-platform build proof (`gix`).
 2. **Clean-build time delta.** Measuring it means building the deps prefix twice, with the
   flag ON and OFF, on the same machine. The incremental figures from day-to-day work do not
   answer the question and are not offered as if they did.
@@ -1,6 +1,6 @@
 # Mate connectors: aligning with the mainstream CAD systems

-Research date: 2026-08-05. Written against `orca_cad` / `snaporca` at the M8 state
+Research date: 2026-08-05. Written against `orca_cad` / `Snapmaker` at the M8 state
 (`CadDocument.{hpp,cpp}`, `apply_mate`, `datum_frame`, the `Mate` card in `DesignPanel.cpp`).

 **Brief:** align with the mate-connector concept as the main CAD programs actually implement it,
@@ -300,7 +300,7 @@ times.

 **C2 — The roll is unspecified.** Aligning Z leaves one rotation about Z undetermined. Something must
 pin it, and if that something is world-derived, the frame does not rotate with its part. **This
-codebase shipped exactly this bug** (`snaporca-en4`): a face-only connector took Z from the face
+codebase shipped exactly this bug** (`en4`): a face-only connector took Z from the face
 normal but X from `coordsys_x_hint`, a world constant, so Fastened and Slider claimed to lock an
 orientation the frame could not see. Fixed 2026-07-26 by deriving X from the face's own first usable
 edge — but note the fix's own caveat: *"replaying an older document whose face-only connector fed a
@@ -512,7 +512,7 @@ Source of record: `CadDocument.hpp:26,247-252,298-310`; `CadDocument.cpp:1669` (
 **Already aligned — do not "fix" these:** the five types and their DOF; the frame definition (A1);
 Z as the joint axis (A2); superimpose-then-relax (A3); the fixed/moving asymmetry in the data model
 (A5); DOF wording in the type list (A7); free-DOF preservation (A8); right-handed frames under mirror
-(R14); and `snaporca-en4`'s fix, which put roll derivation on the body where it belongs (C2).
+(R14); and `en4`'s fix, which put roll derivation on the body where it belongs (C2).

 **The pattern worth naming: the kernel is in good shape and the concept is under-explained.** Half the
 requirements here are wording and drawing, not geometry. The two real engineering items are R9 (origin
@@ -668,7 +668,7 @@ is a symbol, not a part — it must not shrink with the model. Nothing about tha
 The glyph was therefore implemented and driven on the rig. Screenshots: `g-0*.png`, left in the workspace `artifacts/shots/` and not moved into the repo.
 Five findings, none of which a mock could have produced:

-**F1 — Three axis arms lose to one.** Rendered side by side (`SNAPORCA_GLYPH=A` vs default), the
+**F1 — Three axis arms lose to one.** Rendered side by side (`ORCA_CAD_GLYPH=A` vs default), the
 Onshape-style RGB trio crowds a 22 px disc: the arrowheads are as large as the disc, they bury the
 gold quadrant, and at an oblique angle the three heads pile into a coloured smudge. Worse, **it is
 indistinguishable from the move gizmo and the bed triad**, which are already RGB arrow trios in this
@@ -827,7 +827,7 @@ conversation it arrived in points at "glyph".
 It changes the meaning of every stored document containing a mate. Options: (a) invert and migrate,
 writing `direction=Aligned` where `mate_flip` was false; (b) invert only for new mates and store
 `direction` explicitly from now on. (b) is safer and costs one field. Note this project has taken one
-such semantic hit knowingly before — the `snaporca-en4` fix — and the golden fixture survived, so the
+such semantic hit knowingly before — the `en4` fix — and the golden fixture survived, so the
 migration path is a known quantity. **If G3 (live preview) lands first, this matters much less.**

 **D2 — How far to take origin candidates?** [R9]
@@ -911,7 +911,7 @@ R18's loud refusals carrying the honesty.
 [Joint kinematics — the six lower pairs and their DOF](https://erc-bpgc.github.io/handbook/mechanical/Joint%20Kinematics/) ·
 [ISO 10303-105 — Kinematics (STEP integrated resource)](https://www.iso.org/standard/78589.html)

-**Internal** — `snaporca-en4` (closed 2026-07-26, fixes C2 here) · `CadDocument.cpp:1669`
+**Internal** — `en4` (closed 2026-07-26, fixes C2 here) · `CadDocument.cpp:1669`
 `datum_frame` · `CadDocument.cpp:2961` `apply_mate` · `CadDocument.cpp:1302` `add_mate`

 **Second opinion** — an independent review by Kimi Code (2026-08-05) contributed the
@@ -9,7 +9,7 @@ static const Vec2d kBearOutline[] = {        // 12 verts, RDP eps 0.030, CCW
 static const Vec2d kBearChin[] = {            // the CHIN BAR, flat. The muzzle is relief — see kBearCrest.
    {-0.2682, -0.3578}, {+0.2628, -0.3578}, {+0.2237, -0.1786},
 };
-// {cx, cy, r}: two eyes, then the cheek dot that carries handedness (snaporca-wi3z).
+// {cx, cy, r}: two eyes, then the cheek dot that carries handedness (wi3z).
 static const Vec3d kBearMarks[] = {
    {-0.1997, +0.1760, +0.0590},
    {+0.1947, +0.1760, +0.0590},
@@ -96,7 +96,7 @@
 <div class="wrap">

  <header>
-    <div class="eyebrow">SnapOrca Design · assembly</div>
+    <div class="eyebrow">Orca Design · assembly</div>
    <h1>Mate connector glyph — polarity and verse</h1>
    <p class="sub">
      Onshape's core (disc + roll quadrant + Z arrow) is adopted unchanged because it is proven and
@@ -1,4 +1,4 @@
-"""Emit the simplified bear as a C++ table for the viewport glyph — snaporca-wi3z.
+"""Emit the simplified bear as a C++ table for the viewport glyph — wi3z.

 Everything is normalised to the part's own bounding span and centred, so the renderer scales by
 one radius R in screen pixels and nothing here carries millimetres. Emitting rather than
@@ -79,7 +79,7 @@ L.append("};")
 L.append(f"static const Vec2d kBearChin[] = {{            // the CHIN BAR, flat. The muzzle is relief — see kBearCrest.")
 L.append("    "+", ".join(f"{{{fmt(x)}, {fmt(y)}}}" for x,y in TRI)+",")
 L.append("};")
-L.append("// {cx, cy, r}: two eyes, then the cheek dot that carries handedness (snaporca-wi3z).")
+L.append("// {cx, cy, r}: two eyes, then the cheek dot that carries handedness (wi3z).")
 L.append("static const Vec3d kBearMarks[] = {")
 for cx,cy,r in E: L.append(f"    {{{fmt(cx)}, {fmt(cy)}, {fmt(r)}}},")
 L.append(f"    {{{fmt(DOT[0])}, {fmt(DOT[1])}, {fmt(DOT[2])}}},")
@@ -1,4 +1,4 @@
-"""Render the SIMPLIFIED glyph exactly as render_mate_face() draws it — snaporca-x0kd.
+"""Render the SIMPLIFIED glyph exactly as render_mate_face() draws it — x0kd.

 This is the panel the study was missing. simplify_study.py measured a FLAT outline and
 relief_sheet.py measured the FULL 1508-facet part; neither showed the simplified glyph WITH its
@@ -1,4 +1,4 @@
-"""Give the bear a handedness mark that survives rasterisation — snaporca-wi3z, Tommaso's call 2.
+"""Give the bear a handedness mark that survives rasterisation — wi3z, Tommaso's call 2.

 The study showed the left/right cue lives in sub-millimetre corner radii and is therefore invisible
 at glyph size: one pixel is 2.6 mm at 32 px. Roll and verse are safe; handedness is not.
@@ -1,4 +1,4 @@
-"""The muzzle has to READ, not just be present — snaporca-wi3z.
+"""The muzzle has to READ, not just be present — wi3z.

 Faithfully scaled, the part's ridge is 11.3 mm on an 83 mm face: 13.6 % of the width. At glyph
 size that is a scratch. A glyph is a symbol, not a scale model, so the question is how much
@@ -1,4 +1,4 @@
-"""Flat glyph vs 3D relief, at the elevations that killed the disc — snaporca-wi3z.
+"""Flat glyph vs 3D relief, at the elevations that killed the disc — wi3z.

 The flat study collapsed at 16 deg because anything drawn IN the connector's plane foreshortens by
 sin(elevation). This renders the SAME bear as its real relief (1508 facets off the supplied male)
@@ -1,4 +1,4 @@
-"""Reduce the bear face to the fewest marks that still read at glyph size — snaporca-wi3z.
+"""Reduce the bear face to the fewest marks that still read at glyph size — wi3z.

 Geometry comes from bear_outline.json, which extract_outline.py pulled off the supplied male
 B-rep's back plate: the outer wire IS the silhouette, the inner wires are the two eyes and the
@@ -1,13 +1,13 @@
 # Rig build traps

-The build rig is two long-lived containers, `snaporca-gui` and `orcacad-gui`, one per fork. Each
-mounts only its fork's build volume (`snaporca_buildcache` / `orcacad_buildcache`) at
+The build rig is two long-lived containers, `snapmaker-gui` and `orcacad-gui`, one per fork. Each
+mounts only its fork's build volume (`snapmaker_buildcache` / `orcacad_buildcache`) at
 `/OrcaSlicer/build`, its fork's `resources/`, and a shots directory — nothing else. They run the
 binary; they do not build it. Rebuild with `scripts/CAD/build-gui.sh`.

 | fork repo | project() | deps image | build volume | GUI container | binary |
 |---|---|---|---|---|---|
-| `snaporca` | `Snapmaker_Orca` | `snaporca-deps` | `snaporca_buildcache` | `snaporca-gui` | `snapmaker-orca` |
+| `Snapmaker` | `Snapmaker_Orca` | `snapmaker-deps` | `snapmaker_buildcache` | `snapmaker-gui` | `snapmaker-orca` |
 | `orca_cad` | `OrcaSlicer` | `orcacad-deps` | `orcacad_buildcache` | `orcacad-gui` | `orca-slicer` |

 `scripts/CAD/build-gui.sh` exists alongside `scripts/CAD/build-gui-incremental.sh` for one reason: it does a
@@ -29,7 +29,7 @@ reports an unknown target, and `orca-slicer` / `OrcaSlicer` have been replaced b

 **Cause.** The GUI image's baked `/OrcaSlicer` tree is the Jun-13 Snapmaker-derived source
 (`project(Snapmaker_Orca)`, executable `snapmaker-orca`). `orcacad-deps` is layered on
-`snaporca-deps`, so even on the mainline fork the baked tree is the other fork's. A `cmake .`
+`snapmaker-deps`, so even on the mainline fork the baked tree is the other fork's. A `cmake .`
 there reconfigures the shared build dir under the wrong project name.

 **Fix.** Build only via `scripts/CAD/build-gui.sh`, which starts a throwaway container from the deps
@@ -64,10 +64,10 @@ Did you initialize submodules?` (the `FATAL_ERROR` guarding `PYBIND11_SOURCE_DIR
 fork's root `CMakeLists.txt`, near line 948).

 **Cause.** The deps image predates that requirement. Only the mainline (`orca_cad`) fork has
-`deps_src/pybind11` and the requirement; snaporca has neither.
+`deps_src/pybind11` and the requirement; Snapmaker has neither.

 **Fix.** Mount `deps_src` over the baked tree — `scripts/CAD/build-gui.sh` does. Corollary: mounting a
-snaporca tree into an `orcacad-deps` build reproduces this error exactly.
+Snapmaker tree into an `orcacad-deps` build reproduces this error exactly.

 ---

@@ -96,7 +96,7 @@ you ever configure by hand, run it twice.
 **Cause.** The cache carries `SLIC3R_CAD=ON`, but the root `CMakeLists.txt` actually configured is
 a stale baked copy that predates the gate and never runs `add_definitions(-DSLIC3R_CAD)` (the
 gate is `if (SLIC3R_CAD)` / `add_definitions(-DSLIC3R_CAD)` in the root list — line 179/180 in
-snaporca, 319/320 in orca_cad). Every `#ifdef SLIC3R_CAD` block therefore compiles out while the
+Snapmaker, 319/320 in orca_cad). Every `#ifdef SLIC3R_CAD` block therefore compiles out while the
 option still reads ON.

 **Fix.** Always mount the live `CMakeLists.txt` and `cmake/` — never inherit them from the image.
@@ -109,12 +109,12 @@ all mount both.

 `ninja <target>` writes `/OrcaSlicer/build/src/Release/<binary>`; only `build_linux.sh`
 additionally packages to `/OrcaSlicer/build/package/bin/<binary>`. `orca_cad`'s
-`scripts/CAD/start-headless-gui.sh` defaults `BIN` to `src/Release/orca-slicer`, but snaporca's defaults to
-`package/bin/snapmaker-orca`. So after a target-only rebuild on snaporca, launching
+`scripts/CAD/start-headless-gui.sh` defaults `BIN` to `src/Release/orca-slicer`, but Snapmaker's defaults to
+`package/bin/snapmaker-orca`. So after a target-only rebuild on Snapmaker, launching
 `start-headless-gui.sh` with its default runs the **stale packaged** binary — the change under test is
 invisible and the session hunts a phantom. Pass `BIN` explicitly:

-    docker exec -e BIN=/OrcaSlicer/build/src/Release/snapmaker-orca snaporca-gui /OrcaSlicer/scripts/CAD/start-headless-gui.sh
+    docker exec -e BIN=/OrcaSlicer/build/src/Release/snapmaker-orca snapmaker-gui /OrcaSlicer/scripts/CAD/start-headless-gui.sh

 `scripts/CAD/build-gui.sh` prints the correct line for the current fork when it finishes.

@@ -121,7 +121,7 @@ def main():
        # A verb may carry a NOTE: the reason it exists, emitted as a C++ comment above its row.
        # Without somewhere to put it, a rationale written into the generated header is deleted by
        # the next regeneration — which is how the model-mode "Constrain sketch" row came to exist
-        # in the header and not in the atlas at all (snaporca-ziam). The map exists once; so does
+        # in the header and not in the atlas at all (ziam). The map exists once; so does
        # the explanation.
        for ln in ([v["note"]] if isinstance(v.get("note"), str) else v.get("note") or []):
            lines.append(f"    // {ln}")
@@ -23,7 +23,7 @@ The verb in the name is the role:
 | `check-sketch-engine-corpus.py` | The same ladder graded against a systematic sample of real drawings instead of shapes we chose. | Kernel + corpus |
 | `check-gui-sketching.py` | The same profiles drawn the way a person draws them — synthetic mouse gestures and typed values. | Headless GUI |
 | `check-gui-context-menu.py` | That right-click is the pivot of the design gesture, and adapts to what was clicked. | Headless GUI |
-| `check-mcp-sketch.py` | The sketch layer driven over the MCP socket, asserting what decides whether a profile is buildable. | Headless GUI + `SNAPORCA_MCP` |
+| `check-mcp-sketch.py` | The sketch layer driven over the MCP socket, asserting what decides whether a profile is buildable. | Headless GUI + `ORCA_CAD_MCP` |

 **`run-kernel-tests.sh` is the only one CI can run.** The rest need a live
 application with an OpenGL canvas and synthetic input, which hosted runners do not
@@ -19,7 +19,7 @@
 set -euo pipefail

 REPO="$(cd "$(dirname "${BASH_SOURCE[0]}")/../.." && pwd)"
-# orcacad-deps, NOT snaporca-deps: see the note in run-kernel-tests.sh — the wrong image
+# orcacad-deps, NOT snapmaker-deps: see the note in run-kernel-tests.sh — the wrong image
 # fails at CMake configure, not at link time.
 IMAGE="${IMAGE:-orcacad-deps}"
 BUILD_VOL="${BUILD_VOL:-orcacad_buildcache}"
@@ -32,11 +32,11 @@ echo "REPO=$REPO  IMAGE=$IMAGE  BUILD_VOL=$BUILD_VOL"
 # the build fails with "class GLCanvas3D has no member named set_design_sketch_tool".
 #
 # build_linux.sh must be mounted for the same reason, and here the stale copy is guaranteed
-# wrong rather than merely risky: orcacad-deps is layered on snaporca-deps, so the baked script
+# wrong rather than merely risky: orcacad-deps is layered on snapmaker-deps, so the baked script
 # is the OTHER fork's and builds `--target Snapmaker_Orca`. This fork's target is `OrcaSlicer`,
 # so without this mount configure succeeds and then ninja dies on "unknown target".
 # scripts/ likewise: build_linux.sh's packaging step sources scripts/appimage_lib_policy.sh,
-# which the baked snaporca tree does not have, so a fully successful link still exited
+# which the baked Snapmaker tree does not have, so a fully successful link still exited
 # non-zero with "missing AppImage helper" and the binary check never ran.

 # ---- OOM guard (2026-08-21) -------------------------------------------------------------
@@ -2,7 +2,7 @@
 # Rebuild the GUI binary the design rig launches — in a THROWAWAY container, writing into the
 # same build-cache volume the rig's long-lived GUI container reads from.
 #
-# NEVER build inside the GUI container (snaporca-gui / orcacad-gui). Its baked /OrcaSlicer tree
+# NEVER build inside the GUI container (snapmaker-gui / orcacad-gui). Its baked /OrcaSlicer tree
 # is the Jun-13 Snapmaker-derived source, so a `cmake .` in there silently reconfigures the
 # shared build dir as project(Snapmaker_Orca) and this fork's targets vanish. That is Trap 1 of
 # five; all of them, with symptoms and exact recovery commands, are in docs/rig_build_traps.md.
@@ -21,7 +21,7 @@ REPO="$(cd "$(dirname "${BASH_SOURCE[0]}")/../.." && pwd)"
 # the wrong volume the two forks silently trade build artefacts.
 PROJECT="$(sed -n 's/^project(\([A-Za-z_0-9]*\)).*/\1/p' "$REPO/CMakeLists.txt" | head -1)"
 case "$PROJECT" in
-    Snapmaker_Orca) PREFIX=snaporca; BIN=snapmaker-orca ;;
+    Snapmaker_Orca) PREFIX=snapmaker; BIN=snapmaker-orca ;;
    OrcaSlicer)     PREFIX=orcacad;  BIN=orca-slicer    ;;
    *) echo "FATAL: unrecognised project($PROJECT) in $REPO/CMakeLists.txt" >&2; exit 2 ;;
 esac
@@ -14,7 +14,7 @@ again. A user cannot be told to click the field first; when they do not, they ge
 number and report "the label value is not editable". So this ladder types IMMEDIATELY after the
 field opens, exactly as a person does, and fails if the prefill is what gets committed.

-WHAT IT GRADES. The app emits one line per event under SNAPORCA_UXTRACE=1:
+WHAT IT GRADES. The app emits one line per event under ORCA_CAD_UXTRACE=1:

    [UX] open    title=Length prefill=154.76
    [UX] commit  title=Length typed=80 value=80.0000
@@ -29,7 +29,7 @@ that matters — a field that is on screen but deaf commits its prefill, and eve
    scripts/CAD/check-gui-click-edit.py --display :10 --bin build/src/Release/orca-slicer

 With --attach it drives an already-running app instead of launching one; the app must have been
-started with SNAPORCA_UXTRACE=1 and its stderr redirected to --trace.
+started with ORCA_CAD_UXTRACE=1 and its stderr redirected to --trace.
 Exit 0 = every field took what was typed.
 """
 import argparse, json, os, re, shutil, signal, socket, subprocess, sys, tempfile, time
@@ -175,7 +175,7 @@ def launch():
    # the process was alive, `xdotool search` on the rig display found nothing, and the window
    # was sitting on the user's own screen. Silent, and it drives a stray app at someone's face.
    env.pop("WAYLAND_DISPLAY", None)
-    env.update(DISPLAY=DISP, GDK_BACKEND="x11", SNAPORCA_UXTRACE="1",
+    env.update(DISPLAY=DISP, GDK_BACKEND="x11", ORCA_CAD_UXTRACE="1",
               LIBGL_ALWAYS_SOFTWARE="1", GALLIUM_DRIVER="llvmpipe",
               # The rig's Xvfb has no input-method daemon, and a dead ibus context makes a
               # GtkEntry drop every character while the app looks fine. It cannot affect the
@@ -186,7 +186,7 @@ def launch():
               # the Design panel at all. Without it "the field never opened" is indistinguishable
               # from "we never got into sketch mode", and the first run of this ladder reported
               # seven product failures that were really one driver racing a still-loading app.
-               SNAPORCA_KEYTRACE="1", SNAPORCA_MCP=A.sock,
+               ORCA_CAD_KEYTRACE="1", ORCA_CAD_MCP=A.sock,
               SSL_CERT_FILE="/etc/ssl/certs/ca-certificates.crt",
               WEBKIT_DISABLE_DMABUF_RENDERER="1", WEBKIT_DISABLE_COMPOSITING_MODE="1")
    if os.path.exists(A.sock):
@@ -11,16 +11,16 @@ action and no shortcut, so a key-driven ladder cannot reach them at all.
 This ladder drives the menu. Nothing here is asserted from pixels:

  WHAT WAS CLICKED  -> the offer's own [OFFER] trace, emitted by show_offer_menu from the same
-                       loop that builds the rows (SNAPORCA_KEYTRACE). It cannot drift from what
+                       loop that builds the rows (ORCA_CAD_KEYTRACE). It cannot drift from what
                       the user is shown, which a hand-written expectation list would.
  WHAT IS OFFERED   -> the same trace, compared against DesignOffer.hpp parsed independently.
                       "The menu shows exactly the verbs the table says apply here" is a
                       property; a copied list of row names is a transcription.
  WHAT IT PRODUCED  -> the MCP socket, read-only, exactly as in the gesture ladder.

-Run inside the rig container, with the app launched under SNAPORCA_KEYTRACE=1:
+Run inside the rig container, with the app launched under ORCA_CAD_KEYTRACE=1:

-    docker exec snaporca-gui python3 /OrcaSlicer/scripts/CAD/check-gui-context-menu.py [rung ...]
+    docker exec orcacad-gui python3 /OrcaSlicer/scripts/CAD/check-gui-context-menu.py [rung ...]
 """
 import importlib.util
 import math
@@ -38,13 +38,13 @@ _spec = importlib.util.spec_from_file_location("gui_ladder", os.path.join(HERE,
 G = importlib.util.module_from_spec(_spec)
 _spec.loader.exec_module(G)

-LOG = os.environ.get("SNAPORCA_GUI_LOG", "/tmp/gui-session.log")
+LOG = os.environ.get("ORCA_CAD_GUI_LOG", "/tmp/gui-session.log")

 # The rig container's /OrcaSlicer is the image's own baked source tree, not this checkout, so the
 # generated header is not where a repo-relative path expects it. Look in both places and say which
 # one was read — a ladder that silently graded against the WRONG table would be worse than one
 # that refuses to start.
-HEADER_CANDIDATES = [os.environ.get("SNAPORCA_OFFER_HPP", ""),
+HEADER_CANDIDATES = [os.environ.get("ORCA_CAD_OFFER_HPP", ""),
                     os.path.join(HERE, "..", "src", "slic3r", "GUI", "CAD", "DesignOffer.hpp"),
                     os.path.join(HERE, "DesignOffer.hpp")]

@@ -61,12 +61,12 @@ def load_table():
    """Every verb in DesignOffer.hpp, as dicts. The independent half of the comparison.

    Parsed from the generated header rather than from tool_atlas.json on purpose: the header is
-    what the binary was compiled from, and the two have been out of step before (snaporca-ziam,
+    what the binary was compiled from, and the two have been out of step before (ziam,
    where regenerating the header silently dropped the Constrain row).
    """
    path = next((p for p in HEADER_CANDIDATES if p and os.path.exists(p)), None)
    if path is None:
-        raise SystemExit("no DesignOffer.hpp found; set SNAPORCA_OFFER_HPP or copy it beside "
+        raise SystemExit("no DesignOffer.hpp found; set ORCA_CAD_OFFER_HPP or copy it beside "
                         "this script (tried: " + ", ".join(filter(None, HEADER_CANDIDATES)) + ")")
    print(f"offer table: {os.path.realpath(path)}")
    src = open(path).read()
@@ -301,7 +301,7 @@ def rung_kinds():
    # Empty space first: nothing is selected, so the sketch vocabulary's no-selection row set.
    # Right-click has two jobs on a draw tool, and which one it does depends on whether an
    # anchor is down. Both are asserted here: the version that consumed EVERY right-click made
-    # the offer unreachable from any armed tool (snaporca-ghcz), which is the goal's own
+    # the offer unreachable from any armed tool (ghcz), which is the goal's own
    # mechanism failing silently.
    hi = y1 - (y1 - y0) * 0.12
    o = open_offer(cx, hi)
@@ -766,7 +766,7 @@ def rung_curves():
                    f"and it takes a typed radius exactly: {got:.9f} (asked 30.0)")
            # The DoF of ONE CIRCLE is three. Asserted here because it is where the lie showed:
            # after a delete the solver was never re-run, so this reported the DoF of the
-            # geometry that had just been erased. snaporca-ua9g.
+            # geometry that had just been erased. ua9g.
            G.check("VERTEX", G.describe()["dof"] == 2,
                    f"and the sketch reports the DoF of what is actually in it: {G.describe()['dof']}")

@@ -1198,7 +1198,7 @@ def main():
    if not os.path.exists(LOG):
        G.die(f"no {LOG} — launch the app through scripts/CAD/start-headless-gui.sh")
    if "[OFFER]" not in open(LOG, errors="replace").read()[-400000:]:
-        print(f"note: no [OFFER] lines in {LOG} yet — the app must run with SNAPORCA_KEYTRACE=1")
+        print(f"note: no [OFFER] lines in {LOG} yet — the app must run with ORCA_CAD_KEYTRACE=1")
    want = sys.argv[1:] or list(RUNGS)
    # TWICE. From a cold launch the app shows the Home page over the Design tab, and the first
    # click only selects the tab — the second is what brings the viewport forward. A ladder that
@@ -9,16 +9,16 @@ cannot say the Design tab meets its goal. This one draws with synthetic clicks a
 values into the in-canvas field, then reads the result back through the socket, which is used
 here ONLY as an instrument, never as an author.

-Runs INSIDE the headless rig container (Xvfb :10 + openbox + the app with SNAPORCA_MCP set):
+Runs INSIDE the headless rig container (Xvfb :10 + openbox + the app with ORCA_CAD_MCP set):

-    docker cp scripts/CAD/check-gui-sketching.py snaporca-gui:/tmp/ && \
-    docker exec snaporca-gui python3 /tmp/check-gui-sketching.py [rung ...]
+    docker cp scripts/CAD/check-gui-sketching.py orcacad-gui:/tmp/ && \
+    docker exec orcacad-gui python3 /tmp/check-gui-sketching.py [rung ...]

 With no arguments every rung runs. Exit 0 = every property held.
 """
 import json, math, os, re, socket, subprocess, sys, time

-SOCK  = os.environ.get("SNAPORCA_MCP", "/tmp/mcp.sock")
+SOCK  = os.environ.get("ORCA_CAD_MCP", "/tmp/mcp.sock")
 DISP  = os.environ.get("DISPLAY", ":10")
 _n = 0
 _fail = 0
@@ -225,7 +225,7 @@ def leave_sketch():
 # "no sketch opened after plane click + Shift+S"; the unshifted Construction checkbox reported
 # "0 construction axis". Both name the wrong subsystem. Canvas coordinates are immune because
 # clickmm() derives them from the live canvas geometry — only the chrome constants need this.
-CHROME_DY = int(os.environ.get("SNAPORCA_CHROME_DY", "26"))
+CHROME_DY = int(os.environ.get("ORCA_CAD_CHROME_DY", "26"))

 DESIGN_TAB = (128, 29 + CHROME_DY)

@@ -1873,7 +1873,7 @@ def rung_scale():
          f"every cut-out is exactly {side:.6f} squared")
    # Now the part that matters: draw ONE more entity by hand, on top of all that.
    #
-    # No Escape here, deliberately: this rung is the regression test for snaporca-j7gc, where a
+    # No Escape here, deliberately: this rung is the regression test for j7gc, where a
    # bulk sketch_add made while a creation tool is armed was read as a drawn gesture, opened that
    # tool's value field and swallowed the next key and click until one Escape dismissed it. The
    # gesture below has to land on the FIRST try. Fixed by resyncing m_autoedit_seen in
@@ -8,7 +8,7 @@ cost a GUI session and a human. The socket verbs make each one a call, and this
 loop: build a known profile, ask the app what it thinks it has, compare against arithmetic.

 RUN IT AGAINST A RUNNING APP:
-    SNAPORCA_MCP=/tmp/mcp.sock <binary>          # launch with the socket enabled
+    ORCA_CAD_MCP=/tmp/mcp.sock <binary>          # launch with the socket enabled
    python3 scripts/CAD/check-mcp-sketch.py [socket] # default /tmp/mcp.sock

 Exit 0 = every assertion held. Anything else prints the first mismatch and stops.
@@ -35,7 +35,7 @@ import sys
 import tempfile
 import time

-SOCK = os.environ.get("SNAPORCA_MCP", "/tmp/mcp.sock")
+SOCK = os.environ.get("ORCA_CAD_MCP", "/tmp/mcp.sock")
 TOL = 1e-6          # exact-comparison tolerance (all inputs are lines)
 WELD = 0.05         # endpoint-coincidence tolerance, in PDF units

@@ -295,7 +295,7 @@ def grade(pdf, name, report):
        # same rule the engine now uses (DesignSketchTool::region_loops). A vertex is exactly
        # where two loops touch in a real drawing, and a ray cast from a point lying ON the
        # polygon under test answers by rounding: that alone accounted for every one of the 6
-        # sheets where the two attributions used to disagree. snaporca-5hvl.
+        # sheets where the two attributions used to disagree. 5hvl.
        probes = [interior_point(r) for r in rings]
        mine_parent = {}
        for i, r in enumerate(rings):
@@ -15,7 +15,7 @@ Every rung asserts those. Area appears only as a cross-check, never as the verdi

 Entirely 2D: sketch entities only, no extrude, revolve or any solid feature.

-    SNAPORCA_MCP=/tmp/mcp.sock <binary>
+    ORCA_CAD_MCP=/tmp/mcp.sock <binary>
    python3 scripts/CAD/check-sketch-engine.py [socket]

 Exit 0 = every rung held. Otherwise the first broken property is named and the run stops.
@@ -10,7 +10,7 @@
 #   FULL=1 scripts/CAD/run-all-checks.sh          # corpus over ALL 997 sheets (~25 min)
 #   SKIP_GUI=1 scripts/CAD/run-all-checks.sh      # kernel only, for a machine with no rig
 #
-# The rig container is expected to be up with the app running and SNAPORCA_MCP set; bring it up
+# The rig container is expected to be up with the app running and ORCA_CAD_MCP set; bring it up
 # with scripts/CAD/start-headless-gui.sh inside it. The corpus lives at /corpus in that container.
 set -uo pipefail
 # ../.. -- this script lives in scripts/CAD/, so one level up is scripts/, not the repo
@@ -19,7 +19,7 @@ set -uo pipefail
 # scripts/scripts/ and reporting instant failures that were all the same typo.
 cd "$(dirname "${BASH_SOURCE[0]}")/../.." || exit 1

-# orcacad-gui, NOT snaporca-gui: that is the other fork's rig, and defaulting to it makes
+# orcacad-gui, NOT snapmaker-gui: that is the other fork's rig, and defaulting to it makes
 # this gate verify the wrong fork's binary while reporting green. run-kernel-tests.sh
 # carries the same warning about the build volume, where the defect was found first.
 C="${C:-orcacad-gui}"
@@ -53,13 +53,13 @@ run_in_rig() {                      # copy the script in fresh, then run it ther
 # FIRST, and it needs no rig: the offer table the menu is compiled from must be what the atlas
 # says. The header calls itself GENERATED and had been hand-edited anyway — which cost four rows
 # that existed only in the header, one row wired to the wrong action, and a count of 91 for a
-# 92-row array, so the last verb was unreachable (snaporca-z8rs, snaporca-ziam).
+# 92-row array, so the last verb was unreachable (z8rs, ziam).
 # docs/CAD/, not docs/: SoftFever moved the design docs into the CAD subfolder
 # (bbd1989e1e) and this line kept the old path, so the rung failed on a missing file
 # rather than on anything about the table. The other fork still has docs/ux/.
 step "offer table matches the atlas" python3 docs/CAD/ux/mockups/gen_offer_table.py --check

-step "kernel suite" scripts/CAD/run-kernel-tests.sh --vol "${KVOL:-snaporca_kerneltest}"
+step "kernel suite" scripts/CAD/run-kernel-tests.sh --vol "${KVOL:-orcacad_kerneltest}"

 if [ -z "${SKIP_GUI:-}" ]; then
    step "engine ladder (rungs 1-8, scripted geometry)" \
@@ -71,7 +71,7 @@ if [ -z "${SKIP_GUI:-}" ]; then
    step "gesture ladder (mouse and keyboard)" \
        run_in_rig scripts/CAD/check-gui-sketching.py /tmp/check-gui-sketching.py
    # The offer ladder needs TWO extra things the others do not: the app must have been launched
-    # with SNAPORCA_KEYTRACE=1 (its [OFFER] lines are the whole instrument), and it reads the
+    # with ORCA_CAD_KEYTRACE=1 (its [OFFER] lines are the whole instrument), and it reads the
    # generated offer table to predict what each selection should show — which is not in the
    # container's own baked source tree, so it is copied in beside the script — /tmp, where
    # run_in_rig puts the script, is one of the paths the ladder looks in.
@@ -23,18 +23,18 @@
 set -euo pipefail

 REPO="$(cd "$(dirname "${BASH_SOURCE[0]}")/../.." && pwd)"
-# orcacad-deps, NOT snaporca-deps: this fork is mainline-based and needs Eigen 5.0.1,
-# CGAL 5.6.3, wx 3.3.2 and Python 3.12 Development.Embed, none of which snaporca-deps has.
+# orcacad-deps, NOT snapmaker-deps: this fork is mainline-based and needs Eigen 5.0.1,
+# CGAL 5.6.3, wx 3.3.2 and Python 3.12 Development.Embed, none of which snapmaker-deps has.
 # With the wrong image CMake dies at configure, which is exactly why this fork went
 # M1-M8 without ever compiling (see commit 1633005bba).
 IMAGE="${IMAGE:-orcacad-deps}"
-# Must NOT default to snaporca_buildcache: that is the other fork's volume, and pointing
+# Must NOT default to snapmaker_buildcache: that is the other fork's volume, and pointing
 # this fork at it makes the two silently trade build artefacts. build-gui-incremental.sh had the
 # identical defect and was fixed to orcacad_buildcache; this script was missed.
 VOL="${BUILD_VOL:-orcacad_kerneltest}"
 # No exclusions. Both cases that used to be quarantined now run: the solver SIGABRT on
-# circle-line tangency is fixed (snaporca-tkz), and the internal-thread case turned out to have
-# correct geometry and a wrong reference in the test (snaporca-kzy). A green run here now means
+# circle-line tangency is fixed (tkz), and the internal-thread case turned out to have
+# correct geometry and a wrong reference in the test (kzy). A green run here now means
 # the whole CAD suite passed, not "everything except the two we gave up on".
 #
 # ...and that claim was still not true, because the default tag was [CadDocument] alone while
@@ -36,8 +36,8 @@ export LIBGL_ALWAYS_SOFTWARE=1 GALLIUM_DRIVER=llvmpipe
 # them comes up looking perfectly healthy: the window is there, status says app up, and every
 # ladder then dies on "Connection refused" — which reads as a dead app rather than a rig that was
 # started without its instrument. The script that launches the rig is where they belong.
-export SNAPORCA_MCP="${SNAPORCA_MCP:-/tmp/mcp.sock}"
-export SNAPORCA_KEYTRACE="${SNAPORCA_KEYTRACE:-1}"
+export ORCA_CAD_MCP="${ORCA_CAD_MCP:-/tmp/mcp.sock}"
+export ORCA_CAD_KEYTRACE="${ORCA_CAD_KEYTRACE:-1}"
 export LD_LIBRARY_PATH="$LIBPY:$LIBPY2:${LD_LIBRARY_PATH:-}"
 mkdir -p /root/.config          # startup dies in boost::filesystem::create_directory without this

@@ -1,13 +1,13 @@
-# Deps-only base image for fast iteration on SnapOrca.
+# Deps-only base image for fast iteration on Orca.
 # Identical system+pinned-dependency setup to scripts/Dockerfile, but STOPS after
 # `build_linux.sh -dr` (no slicer/AppImage build). Produces an image with the pinned
 # deps baked at /OrcaSlicer/deps/build/destdir, so the slicer can be rebuilt
 # incrementally via scripts/CAD/build-gui-incremental.sh without re-running the long deps build.
 #
 # Build once (rebuild only when deps/ changes, e.g. OCCT module flags):
-#   docker build -t snaporca-deps -f scripts/Dockerfile.deps .
+#   docker build -t snapmaker-deps -f scripts/Dockerfile.deps .
 FROM docker.io/ubuntu:24.04
-LABEL maintainer="SnapOrca CAD iteration base"
+LABEL maintainer="Orca CAD iteration base"

 # Disable interactive package configuration
 RUN apt-get update && \
@@ -80,11 +80,11 @@ RUN ./build_linux.sh -dr -j 12
 RUN ln -sfn /OrcaSlicer/deps/build/OrcaSlicer_dep /OrcaSlicer/deps/build/destdir

 # The rig's GUI runtime. This used to arrive for free because orcacad-deps was layered on
-# snaporca-deps; that lineage is Trap 1 in docs/rig_build_traps.md (a baked project(Snapmaker_Orca)
+# snapmaker-deps; that lineage is Trap 1 in docs/rig_build_traps.md (a baked project(Snapmaker_Orca)
 # tree) and building from this Dockerfile is what removes it — along with the X stack the rig
 # needs. scripts/CAD/start-headless-gui.sh requires Xvfb and openbox (without a window manager `xdotool
 # windowactivate` aborts with "windowmanager claims not to support..."), drives the UI with
-# xdotool, and captures to /shots with scrot/ImageMagick. Same set snaporca-deps carries.
+# xdotool, and captures to /shots with scrot/ImageMagick. Same set snapmaker-deps carries.
 RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \
    imagemagick \
    openbox \
@@ -9,8 +9,8 @@
 #
 # The consequence was that scripts/CAD/run-kernel-tests.sh failed at CMake CONFIGURE time,
 # before a single source file compiled, so THIS FORK'S KERNEL SUITE COULD NOT RUN AT ALL.
-# Every kernel change ported here was parity-checked against snaporca and never independently
-# tested (snaporca-w80c). The Snapmaker fork does not hit this: its base requires neither
+# Every kernel change ported here was parity-checked against Snapmaker and never independently
+# tested (w80c). The Snapmaker fork does not hit this: its base requires neither
 # assimp nor OpenCV.
 #
 # WHY A LAYER AND NOT A FULL DEPS REBUILD. Rebuilding every dependency takes hours and would
@@ -325,7 +325,7 @@ void AppConfig::set_defaults()
    // Design tab: draw a mate connector as a face rather than as the abstract disc + roll
    // quadrant. Defaults ON — face orientation is hardwired perception, so the roll and the
    // verse read without being learned, which no abstract glyph achieves. Turning it off
-    // restores the conventional CAD representation for users who expect it (snaporca-x0kd).
+    // restores the conventional CAD representation for users who expect it (x0kd).
    if (get("design_connector_face_glyph").empty())
        set_bool("design_connector_face_glyph", true);
 #endif
@@ -2142,7 +2142,7 @@ TopoDS_Wire CadDocument::build_sketch_wire(const CadFeature& sketch, bool closed
        // legacy tail of this function ends in a default rectangle built from width/height,
        // which for an entity sketch are whatever they happened to be initialised to — so a
        // sketch entities_to_wire cannot handle (a circle coexisting with a line, two circles:
-        // snaporca-88v) used to extrude into a box the user never drew, silently. Failing here
+        // 88v) used to extrude into a box the user never drew, silently. Failing here
        // costs the caller an error message; falling through cost them wrong geometry that
        // looked deliberate. The legacy profile/shape paths below are still reached by sketches
        // that legitimately carry no entities at all.
@@ -2751,7 +2751,7 @@ void CadDocument::apply_feature(TopoDS_Shape& result, bool& have_body,
        // Neutral plane = horizontal plane through the body's bbox bottom, pull direction +Z.
        // The face pivots about the line where it meets the neutral plane and tilts by the angle.
        // ponytail: neutral plane / pull direction fixed to world up; pick-based neutral plane
-        // deferred (same as the datum-plane pick types, snaporca-dgv).
+        // deferred (same as the datum-plane pick types, dgv).
        Bnd_Box bb; BRepBndLib::Add(result, bb);
        Standard_Real xmin, ymin, zmin, xmax, ymax, zmax;
        bb.Get(xmin, ymin, zmin, xmax, ymax, zmax);
@@ -3537,7 +3537,7 @@ void CadDocument::apply_mate(std::vector<CadBody>& bodies, const CadFeature& f)
 }

 // Volume of a shape, 0 for anything that isn't a solid we can measure. Used to catch a
-// subtraction that removed nothing (snaporca-daf).
+// subtraction that removed nothing (daf).
 static double solid_volume(const TopoDS_Shape& s)
 {
    if (s.IsNull()) return 0.0;
@@ -3591,7 +3591,7 @@ void CadDocument::route_feature(std::vector<CadBody>& bodies, const CadFeature&
        // an agent especially — then has no signal at all that the hole it asked for was never
        // drilled: same body, same volume, ok:true. Measure the volume across the op and refuse
        // the no-op. Only for removals: every other feature type may legitimately leave the volume
-        // alone (a Transform certainly does). snaporca-daf.
+        // alone (a Transform certainly does). daf.
        const bool removes = f.type == CadFeatureType::Hole
                          || f.type == CadFeatureType::Thread
                          || ((f.type == CadFeatureType::Extrude || f.type == CadFeatureType::Revolve
@@ -3625,7 +3625,7 @@ bool CadDocument::recompute()
    // made a sketch-only design unsaveable AND unopenable: DesignPanel::recompute_guarded syncs
    // the 3MF recipe only "on success", so nothing was written, and deserialize_recipe ends with
    // `return recompute()`, so a project that did carry a recipe was refused on load with
-    // "Could not restore the CAD model" while its features sat correctly in the list. snaporca-mtav.
+    // "Could not restore the CAD model" while its features sat correctly in the list. mtav.
    bool any_solid_feature = false;
    try {
        // Parametric pass: evaluate document variables, then each feature's expression bindings,
@@ -403,7 +403,7 @@ static SketchSolveResult solve_system(std::vector<SketchEntity>& entities,
    // from the corpse. The fillet degrade ladder hit this on every corner — rung 1 (a
    // tangent on each leg) is legitimately over-constrained against the legs' own H/V, and
    // its wreckage then failed rungs 2 and 3, which solve cleanly on their own. The arc
-    // ended up with no constraints at all and the solver snapped the corner shut. snaporca-pl5.
+    // ended up with no constraints at all and the solver snapped the corner shut. pl5.
    if (!out.ok) return out;
    for (size_t i = 0; i < entities.size(); ++i) {
        SketchEntity& e = entities[i];
@@ -97,7 +97,7 @@ DesignCanvas::DesignCanvas(wxWindow* parent)
    // this point is known good (ImGui reports want_text=1 and our InputText active), so if these
    // lines do not appear the character never got past the panel's CHAR_HOOK / the focus chain,
    // and no amount of work inside the field will help. Skips always: a pure observer.
-    if (m_canvas_widget != nullptr && std::getenv("SNAPORCA_UXTRACE")) {
+    if (m_canvas_widget != nullptr && std::getenv("ORCA_CAD_UXTRACE")) {
        m_canvas_widget->Bind(wxEVT_CHAR, [](wxKeyEvent& e) {
            fprintf(stderr, "[UX] canvas_char key=%d\n", e.GetKeyCode());
            fflush(stderr);
@@ -1240,7 +1240,7 @@ void DesignCanvas::set_status_text(const wxString& text, const wxColour& colour)
 }

 // SetLabel + Wrap + Fit, in that order and always together. Moving the status out of the panel
-// removed the clipping of snaporca-8cc but not the underlying problem: the chip is a top-level
+// removed the clipping of 8cc but not the underlying problem: the chip is a top-level
 // popup that Fit()s to its text, so a long sentence simply grew past the right edge of the canvas
 // and hung over the window. Wrapping to the room actually available is what makes the earlier
 // promise — "a sentence can be a sentence" — true at every window width, including the charter's
@@ -1274,7 +1274,7 @@ void DesignCanvas::place_status_hud()
    const wxPoint bl = m_canvas_widget->ClientToScreen(
        wxPoint(kLeftInset, cs.GetHeight() - hs.GetHeight() - 12));
    // No Raise() and no focus juggling: a popup neither takes focus nor falls behind. This was
-    // caught with SNAPORCA_KEYTRACE — shift+S logged a line, the following R logged nothing, and
+    // caught with ORCA_CAD_KEYTRACE — shift+S logged a line, the following R logged nothing, and
    // the only thing between them was the first status update showing this window.
    if (!m_status_hud->IsShown()) m_status_hud->Show();   // Show before Move (GTK ignores pre-map Move)
    m_status_hud->Move(bl);
@@ -94,7 +94,7 @@ public:
    void set_on_segment_drawn(std::function<void(double, double)> cb);
    void set_on_cursor_metrics(std::function<void(double, double, bool)> cb);
    void set_on_solve_state(std::function<void(int, bool, bool)> cb);  // dof, ok, has_constraints
-    // Live per-step guidance from the armed sketch tool (mode, step, picks). snaporca-1c0c.
+    // Live per-step guidance from the armed sketch tool (mode, step, picks). 1c0c.
    void set_on_sketch_step(std::function<void(DesignSketchTool::Mode, int, int)> cb);
    void apply_segment_length(double len);  // exact length, then commit & repaint
    void keep_segment_as_drawn();           // commit as-drawn & repaint
@@ -219,12 +219,12 @@ public:
    void set_highlight_sketches(std::vector<std::pair<int, ColorRGBA>> hl);
    void set_datum_planes(std::vector<SketchPlane> planes,
                           std::vector<Vec2d> sizes = {});     // draw datum/reference planes (u/v extents)
-    // Mate connectors, drawn as frames so their verse and polarity are visible (snaporca-wgsc).
+    // Mate connectors, drawn as frames so their verse and polarity are visible (wgsc).
    void set_mate_connectors(std::vector<DesignSketchTool::MateConnectorGlyph> g);
    void set_mate_links(std::vector<std::pair<Vec3d, Vec3d>> l);
    void set_body_highlight(bool on);   // tint the solid when its feature is tree-selected
    // The status line, shown along the BASE OF THE VIEWPORT rather than in the side panel:
-    // the panel clips it at ~73 characters with no warning (snaporca-8cc), the viewport's
+    // the panel clips it at ~73 characters with no warning (8cc), the viewport's
    // bottom margin has the whole window width to spare. Empty text hides it.
    void set_status_text(const wxString& text, const wxColour& colour);
    // Take the status line down / bring it back when the Design page leaves and re-enters view.
@@ -62,7 +62,7 @@
 #include "slic3r/GUI/MainFrame.hpp"
 #include "slic3r/GUI/GUI_ObjectList.hpp"

-// English-only pin for the Design tab (see snaporca-design-ux-contract): one lever
+// English-only pin for the Design tab (see design-ux-contract): one lever
 // de-translates this whole TU so our strings never half-translate against the host's
 // localized chrome. Host UI still follows the app locale; only this tab is pinned EN.
 // GOTCHA: every _L(...) in this file must take a STRING LITERAL (FromUTF8 wants const char*).
@@ -452,7 +452,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
        }
    };
    m_keys_sketch['Q'] = [this] {
-        // With geometry selected, Q converts THAT geometry (snaporca-6zic) — the reading
+        // With geometry selected, Q converts THAT geometry (6zic) — the reading
        // everyone arrives with from other sketchers. With nothing selected it keeps its
        // old meaning: arm construction for whatever you draw next.
        if (m_viewport && m_viewport->is_sketching() &&
@@ -547,7 +547,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
    // still BUILT — that is what registers its "fly:<family>#<row>" address and its Shift+key —
    // but it is never placed on the bar. Hiding rather than skipping construction is deliberate:
    // the addresses are created inside the widget-building loops, so not building would silently
-    // delete 42 verbs from the offer while they still rendered. snaporca-7ih records the cleanup
+    // delete 42 verbs from the offer while they still rendered. 7ih records the cleanup
    // that lets the construction go away too.
    // What stays: the two doc-row imports (consumed by add_doc below) and the view controls,
    // which are chrome_only in the atlas and so have no offer row to fall back on.
@@ -577,7 +577,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
            // outright for a family the bar no longer carries. It used to sit INSIDE the build
            // loop, so a retired family still had to be constructed and then Hide()n: skipping it
            // would have deleted 42 verbs from the offer while their rows still rendered and did
-            // nothing when picked. snaporca-7ih.
+            // nothing when picked. 7ih.
            // Keyed on "fly:<family>#<row>" so the generated table can name a variant without the
            // item struct growing a field at 26 call sites.
            for (size_t i = 0; i < vars.size(); ++i) {
@@ -746,7 +746,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
                // Thicken and were then asked to point at something. Reached from the offer the
                // verb is invoked ON a face, so discarding it opened the card reading "(pick a
                // solid face)" over an immediate "thicken: face not found" — the user pointed at
-                // the face and the card said it could not find one. snaporca-kgx.
+                // the face and the card said it could not find one. kgx.
                // The index is per-body, so it only survives if the body combo landed on the body
                // it came from; selected_body_default() above returns exactly that when valid.
                if (m_thicken_body->GetSelection() != m_sel_solid_body)
@@ -1267,7 +1267,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
            // The offer reaches each tool by its ratified address; without these the offer could
            // name a family but only ever arm its FIRST tool: picking "Rectangle" ran key:R and
            // gave you a corner rectangle, with oblique and rounded unreachable. Keyed on the icon
-            // id (already unique per family) so no call site grows an argument. snaporca-6vs.
+            // id (already unique per family) so no call site grows an argument. 6vs.
            for (size_t i = 0; i < vars.size(); ++i) {
                const DesignSketchTool::Mode mode = vars[i].mode;
                const wxString               hint = vars[i].hint;
@@ -1423,7 +1423,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
        // the offer's Create > Polygon submenu. They used to sit inline in this row, then in a
        // sidebar card; both put the choice somewhere you had to leave the geometry to reach,
        // and the count cannot be recovered afterwards (a drawn polygon's inline editor offers
-        // Side and Angle, never the count). snaporca-e1p.
+        // Side and Angle, never the count). e1p.
        auto arm_polygon = [this, select_tool] {
            push_polygon_params();
            select_tool(DesignSketchTool::Mode::Polygon,
@@ -1444,7 +1444,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
        // where you chose the tool — not behind a card you must open to discover it existed.
        // Each address opens the tool exactly as its shortcut does, then says which one.
        // The members are read at INVOCATION, not capture: the cards are built after this row.
-        // snaporca-e1p.
+        // e1p.
        auto open_feature = [this](int key) {
            auto it = m_keys_feature.find(key);
            if (it != m_keys_feature.end() && it->second) it->second();
@@ -1745,7 +1745,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
    // NO plane row. A sketch takes its plane from what is picked in the VIEWPORT — a planar face
    // on a solid, or one of the reference-plane ghosts clicked in 3D — resolved by
    // sketch_plane_from_selection(). A three-row XY/XZ/YZ combo could not express either of those
-    // targets, so it displayed a value that was at best redundant and at worst false. snaporca-e1p.
+    // targets, so it displayed a value that was at best redundant and at worst false. e1p.

    m_width = make_spin(m_cards, 20);
    form->Add(new wxStaticText(m_cards, wxID_ANY, _L("Width / X")), 0, wxALIGN_CENTER_VERTICAL);
@@ -1808,7 +1808,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
        // tool — the offer's Create > Polygon submenu names the common side counts and the two
        // fits, and arming from there sets both. A spin field on the left could not be reached
        // without leaving the geometry, and the count is unrecoverable afterwards: the inline
-        // editor a drawn polygon opens offers Side and Angle, never the count. snaporca-e1p.
+        // editor a drawn polygon opens offers Side and Angle, never the count. e1p.
    }

    // --- Extrude dialog (consumes the selected sketch) ---
@@ -3027,7 +3027,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
        // on a solid, or one of the reference-plane ghosts — because that is where the user is
        // looking and pointing. A combo duplicated that decision somewhere the geometry could not
        // see it, and once a face could be picked it went further and displayed a stale row that
-        // contradicted the real target. snaporca-e1p.
+        // contradicted the real target. e1p.
        // Kept as a member, not a local: the card has to be able to STOP saying this. It asked
        // for a plane even when one had just been picked, directly contradicting the status line
        // two inches below it, which by then read "Sketching on XZ".
@@ -3138,7 +3138,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
    // "slow double-click renames" the old comment promised does not survive wxGTK, which fires
    // ITEM_ACTIVATED first), none of the seven header icons renames, and F2 is a function key
    // nothing announces. A user who wants to name a sketch tries the row, and now the row
-    // answers. snaporca-rename.
+    // answers. rename.
    m_tree->Bind(wxEVT_TREE_ITEM_RIGHT_CLICK, [this](wxTreeEvent& e) {
        m_tree->SelectItem(e.GetItem());          // right-click targets what it points at
        const int sel = tree_selection();
@@ -3622,7 +3622,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
    // Selection no longer writes the status line: on_sketch_step owns it, says the same thing
    // for Select mode and — unlike this callback, which also fired while an edit-op mirrored its
    // picks into the selection — never claims "N selected, Delete removes them" in the middle of
-    // a Mirror gesture, where Delete does nothing of the sort. snaporca-1c0c.
+    // a Mirror gesture, where Delete does nothing of the sort. 1c0c.

    // Onshape flow: clicking inside a closed-loop face commits the sketch and opens
    // the Extrude dialog (with a ghost preview) targeting that sketch.
@@ -3667,7 +3667,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
        // exists to remove. Only a stroke hit carries an entity (an interior click is a region,
        // not a line), so a click inside a loop deliberately leaves the field alone rather than
        // resetting it to something arbitrary. The sketch picker follows the same pick, so
-        // pointing at a line in a different sketch retargets both together. snaporca-3648.
+        // pointing at a line in a different sketch retargets both together. 3648.
        if (m_active == Tool::Rib && entity >= 0) {
            if (m_rib_sketch != nullptr)
                for (unsigned i = 0; i < m_rib_sketch->GetCount(); ++i)
@@ -3741,7 +3741,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
        m_sel_solid_edge = (level == 3) ? edge : -1;
        m_sel_solid_vertex = (level == 4);
        // Keep the hit face even at whole-body level: the cycle's first click means "this body",
-        // but the user pointed AT a face and a sketch should be able to use it. snaporca-3a2.
+        // but the user pointed AT a face and a sketch should be able to use it. 3a2.
        m_pick_face_body = (level >= 1) ? body : -1;
        m_pick_face      = (level >= 1) ? face : -1;
        // Last pick wins: selecting a solid drops any stale committed-sketch loop selection.
@@ -3762,7 +3762,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
        if (m_active == Tool::Dressup) { sync_dressup_target(); update_fillet_gizmo(); refresh_preview(); }
        // Boolean card open: the VIEWPORT is how you choose the two operands. Until now they
        // could only come from two combos — the one control the charter names for this tool
-        // (e1p item 4), and the same pair snaporca-7xx caught silently resolving every row to
+        // (e1p item 4), and the same pair 7xx caught silently resolving every row to
        // index 0. The highlight already flowed card -> viewport; this closes the loop the
        // other way. First pick is the target (kept), second is the tool (consumed); the
        // combos mirror both, so the typed half of L2 still works and still round-trips.
@@ -3787,7 +3787,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
            }
        }
        // Mirror card open: the body you point at is the body that gets mirrored. Same one-way
-        // flow Boolean had (snaporca-310o) and the same fix — the combo stays as the typed half.
+        // flow Boolean had (310o) and the same fix — the combo stays as the typed half.
        // Only one operand here, so there is no slot to alternate and no swap to do.
        if (m_active == Tool::Mirror && m_sel_solid_body >= 0 && m_mirror_body != nullptr &&
            m_sel_solid_body < int(m_mirror_body->GetCount())) {
@@ -3914,12 +3914,12 @@ DesignPanel::DesignPanel(wxWindow* parent)
        m_status->SetForegroundColour(wxNullColour);
        const int nb = int(m_doc.bodies.size());
        const wxString bodytag = (nb > 1) ? wxString::Format(_L("Body %d "), body + 1) : wxString();
-        // Each sub-element line ends by naming the NEXT click (snaporca-gem). Escalation to the
+        // Each sub-element line ends by naming the NEXT click (gem). Escalation to the
        // whole body is a gesture nothing on screen would otherwise reveal, and the status line
        // is the only surface that can teach it at the moment it applies. It REPLACES the old
        // per-level verb hints ("right-click to push/pull it", "Fillet/Chamfer to dress it")
        // rather than joining them: the line is clipped at the panel edge past ~55 characters
-        // (set_status's Wrap() does not take effect — snaporca-8cc), and those verbs are shown
+        // (set_status's Wrap() does not take effect — 8cc), and those verbs are shown
        // with their icons in the offer anyway, while this gesture is shown nowhere else.
        // Both clauses fit now that the line is drawn over the viewport instead of squeezed
        // into the panel. Say "what applies to it", never "verbs" — that is this codebase's
@@ -4173,13 +4173,13 @@ DesignPanel::DesignPanel(wxWindow* parent)
        // select_tool() is what the sketch keys call — so the first letter after entering
        // sketch mode fell through to the feature map, matched nothing (feature keys are
        // Shift+letter), and did nothing. The mouse worked only because the toolbar flyout
-        // reaches select_tool() directly. That is why all 17 keys read as dead. snaporca-0ud.
+        // reaches select_tool() directly. That is why all 17 keys read as dead. 0ud.
        const bool sketch_mode = (m_ui_mode == UiMode::Sketch);
        // Never steal editing keys from a focused text field or an open in-canvas value field —
        // Delete/Ctrl+Z there must edit the text, not the model.
        const bool in_text = (dynamic_cast<wxTextCtrl*>(wxWindow::FindFocus()) != nullptr)
                             || (m_viewport && m_viewport->inline_busy());
-        if (getenv("SNAPORCA_KEYTRACE")) {
+        if (getenv("ORCA_CAD_KEYTRACE")) {
            wxWindow* fw = wxWindow::FindFocus();
            fprintf(stderr, "[KEYTRACE] key=%d ui_mode=%d is_sketching=%d in_text=%d inline_busy=%d focus=%s\n",
                    key, int(m_ui_mode), (m_viewport && m_viewport->is_sketching()) ? 1 : 0, in_text ? 1 : 0,
@@ -4256,7 +4256,7 @@ DesignPanel::DesignPanel(wxWindow* parent)
        // Delete — the selected sketch entities (or the last drawn one if none is selected), or the
        // selected feature in Feature mode. Focus-independent, same reason as undo above.
        // WXK_BACK too: on a keyboard whose Del is a chord (every laptop this runs on), Del is
-        // the one destructive key nobody can reach, and Backspace is what users press. snaporca-oql1.
+        // the one destructive key nobody can reach, and Backspace is what users press. oql1.
        if (!in_text && (key == WXK_DELETE || (key == WXK_BACK && sketching))) {
            if (sketching) { m_viewport->delete_selected_or_last_sketch_entity(); return; }
            if (m_ui_mode == UiMode::Feature && m_active == Tool::None
@@ -4451,7 +4451,7 @@ void DesignPanel::set_ui_mode(UiMode m)
    if (m != UiMode::Sketch) m_sketch_on.clear();   // no stale "on the picked face" on the next hint
    // The DoF readout describes a SKETCH's constraint state, so it means nothing back in Feature
    // mode — where it nonetheless stayed on screen after every Confirm, Cancel and Escape
-    // (snaporca-752). Cleared here rather than at those three exits because this is the one place
+    // (752). Cleared here rather than at those three exits because this is the one place
    // all of them pass through, and a fourth exit added later would otherwise reintroduce it.
    // Constrain mode keeps it: that is where the number is the whole point.
    if (m == UiMode::Feature && m_dof_status != nullptr) {
@@ -4691,7 +4691,7 @@ static void run_off_ui_thread(wxWindow* parent, const wxString& message, const s
 //
 // This used to be written in exactly ONE place — on_commit(), as a side effect of Commit to
 // Plate — so a user who modelled for an hour and pressed Ctrl+S saved a project containing no
-// feature history at all, and the app reported success (snaporca-vjk5). The 3MF exporter was
+// feature history at all, and the app reported success (vjk5). The 3MF exporter was
 // never at fault: nothing had handed it a recipe.
 //
 // Every save path (Ctrl+S, Save As, autosave, crash recovery) reads model.cad_recipe, so
@@ -5080,7 +5080,7 @@ void DesignPanel::on_add_extrude()
    } else if (extrude_uses_loop()) {
        // Extrude just the selected loop (its entity subset), leaving the source sketch's
        // other loops intact and still selectable.
-        if (::getenv("SNAPORCA_PICK_TRACE"))
+        if (::getenv("ORCA_CAD_PICK_TRACE"))
            std::fprintf(stderr, "[pick] on_add_extrude: feat=%d reg=%d ents=%zu\n",
                         m_extrude_sketch_ref, m_sel_sketch_region,
                         m_viewport->selected_loop_entities().size());
@@ -5152,11 +5152,11 @@ void DesignPanel::on_add_dressup()
 // Hole and Thread LATCH the geometry they were opened or picked on; Thicken / Shell / Draft read
 // the live selection instead. Both models are right for what they are — a placement tool with its
 // own plane state, versus an operation whose operand IS the selected face — and the latch is the
-// kinder of the two now that a click on empty canvas clears the selection (snaporca-od0): a stray
+// kinder of the two now that a click on empty canvas clears the selection (od0): a stray
 // click costs a Thicken pick, and costs a Hole nothing. What was missing is that nothing in the
 // Hole/Thread card NAMED the latched face, so after such a click the only words on screen were
 // the viewport's "Nothing selected" — over a ghost still drawn on the face Confirm would drill.
-// That reads as a contradiction and was filed as one (snaporca-200). The card now says what it
+// That reads as a contradiction and was filed as one (200). The card now says what it
 // holds, the way the other three cards already do. Pass -1 for "none, using the plane dropdown".
 void DesignPanel::set_hole_target_label(int face)
 {
@@ -6025,7 +6025,7 @@ SketchPlane DesignPanel::plane_from_choice(int row) const
 // clicking one of the ghost planes in 3D (on_datum_base_picked) rather than by opening the combo.
 // Before this, a picked face was ignored and the only way onto it was to build a Coincident datum
 // plane first and then find it in a dropdown — three steps and a junk feature in the tree for the
-// most common gesture in solid modelling. snaporca-3a2.
+// most common gesture in solid modelling. 3a2.
 SketchPlane DesignPanel::sketch_plane_from_selection(wxString& what) const
 {
    SketchPlane p;
@@ -6080,7 +6080,7 @@ bool DesignPanel::sketch_map_applies() const
 // actually REACHED, so the menu describes what is highlighted — a header that names a face while
 // the whole body is lit would be lying, and this menu's whole value is that it tells the truth
 // about the selection. (Sketching on the face you merely clicked is unaffected: that path is
-// sketch_plane_from_selection, which deliberately uses m_pick_face. snaporca-3a2.)
+// sketch_plane_from_selection, which deliberately uses m_pick_face. 3a2.)
 int DesignPanel::offer_selection_kind() const
 {
    if (sketch_map_applies()) {
@@ -6203,7 +6203,7 @@ void DesignPanel::set_status(const wxString& text)
    // sets the colour on it just before calling here, so this stays the one place that knows
    // both. What the user reads is drawn along the BASE OF THE VIEWPORT: in the panel the line
    // was clipped at ~73 characters with no warning and no wrap (Wrap() never took effect —
-    // snaporca-8cc), which silently length-limited every hint in the tab. The viewport's bottom
+    // 8cc), which silently length-limited every hint in the tab. The viewport's bottom
    // margin has the whole window width, so a sentence can be a sentence.
    if (m_viewport != nullptr) {
        // wxNullColour means "no opinion", and the dark default text colour is nearly invisible
@@ -6218,7 +6218,7 @@ void DesignPanel::set_status(const wxString& text)
 }


-// The sentence for the step the armed sketch tool is on (snaporca-1c0c). One table, so a tool's
+// The sentence for the step the armed sketch tool is on (1c0c). One table, so a tool's
 // gesture is described in one place and the description cannot drift from the code that reads the
 // clicks: the step counts here are the ones DesignSketchTool::render previews and on_mouse
 // consumes. `step` = anchors already placed (edit-ops: 0 none, 1 first pick down, 2 ready to
@@ -6414,7 +6414,7 @@ wxMenuItem* DesignPanel::append_offer_item(wxMenu* menu, int id, const wxString&
 // honest source for that is the loop that builds the rows.
 static void offer_trace(const char* fmt, ...)
 {
-    static const bool on = std::getenv("SNAPORCA_KEYTRACE") != nullptr;
+    static const bool on = std::getenv("ORCA_CAD_KEYTRACE") != nullptr;
    if (!on) return;
    va_list ap;
    va_start(ap, fmt);
@@ -6430,13 +6430,13 @@ void DesignPanel::show_offer_menu(const wxPoint& screen_pos)
    const int      kind = offer_selection_kind();
    const uint32_t bit  = offer_bit(OfferSel(kind));
    // Which verb MAP applies is a question about the MODE, not about whether a session is
-    // running — the same distinction the keyboard already had to learn (snaporca-0ud). Gated on
+    // running — the same distinction the keyboard already had to learn (0ud). Gated on
    // is_sketching() the offer opened on entering a sketch showing the FEATURE rows, every one
    // of them refusing the sketch selection, so it read as a menu of nine dead entries.
    const bool     sketching = sketch_map_applies();
    // The offer ladder reads THIS, not the pixels: the trace is emitted from the same loop that
    // builds the menu, so it cannot drift from what the user is shown. Gated on the existing
-    // SNAPORCA_KEYTRACE so a rig run needs one env var, not two. snaporca-<offer ladder>.
+    // ORCA_CAD_KEYTRACE so a rig run needs one env var, not two. <offer ladder>.
    offer_trace("open kind=%d sketching=%d bodies=%d", kind, sketching ? 1 : 0,
                int(m_doc.bodies.size()));

@@ -6551,7 +6551,7 @@ void DesignPanel::show_offer_menu(const wxPoint& screen_pos)
        }
    }

-    // --- Mate palette section (snaporca-lukg part B) ---
+    // --- Mate palette section (lukg part B) ---
    // Fed by CadDocument::mate_options() so the offer can never disagree with the kernel about
    // which assembly mates a connector pair admits. Shown only when the document holds at least
    // two ENABLED CoordSys features: below that the whole section would be one permanently dead
@@ -7020,7 +7020,7 @@ wxString DesignPanel::idle_hint() const
        : _L("No solid yet — select a sketch and right-click it to Extrude.");
 }

-// The Design tab is no longer the visible page (snaporca-dlj). The status line is a popup floating
+// The Design tab is no longer the visible page (dlj). The status line is a popup floating
 // over the GL canvas, so it does NOT go away when this page does — it stayed up over Prepare and
 // over the home screen, still reading like a live Design selection ("selected (whole body) —
 // right-click for what applies to it") on a tab that has no such selection and no such menu.
@@ -7123,7 +7123,7 @@ void DesignPanel::refresh_tree()
    // recompute() returns FALSE for a document that has no solid ("no solid-producing features",
    // CadDocument.cpp) — which is precisely a document the user has only drawn sketches in. So
    // drawing a profile, pressing Confirm and saving wrote a 3MF with no orca_cad.bin in it
-    // at all, and the app reported success: the whole design was gone on reopen (snaporca-mtav).
+    // at all, and the app reported success: the whole design was gone on reopen (mtav).
    // The three sites that say "a lone sketch yields an empty body; that is expected" call
    // m_doc.recompute() directly and so never reached the sync either. One hook here covers all
    // of them, including the live sketch tool's own commit path.
@@ -9092,7 +9092,7 @@ void DesignPanel::load_feature_into_dialog(const CadFeature& f)
        m_hole_through->SetValue(f.hole_through);
        m_hole_x->SetValue(f.hole_x);
        m_hole_y->SetValue(f.hole_y);
-        // Re-latch the on-face state FROM THE STORED FEATURE (snaporca-uif9). m_hole_on_face is
+        // Re-latch the on-face state FROM THE STORED FEATURE (uif9). m_hole_on_face is
        // only ever cleared by the Hole flyout and by the plane combo, so after any on-face hole
        // it stays true — and a re-edit then drilled on whatever face was latched last, which may
        // be a different face, a different body, or a body since rebuilt. f is the only source
@@ -9123,7 +9123,7 @@ void DesignPanel::load_feature_into_dialog(const CadFeature& f)
        m_thread_x->SetValue(f.thread_x);
        m_thread_y->SetValue(f.thread_y);
        if (m_thread_std) m_thread_std->SetSelection(0);   // Custom: spins reflect the stored feature
-        // Same latch, same failure, same fix as Hole above (snaporca-uif9).
+        // Same latch, same failure, same fix as Hole above (uif9).
        m_thread_on_face    = !is_base_plane(f.plane, m_doc.modeling_origin);
        m_thread_face_plane = f.plane;
        m_thread_face_body  = m_thread_on_face ? f.target_body : -1;
@@ -9702,7 +9702,7 @@ CadFeature DesignPanel::build_candidate(Tool t) const
        // The plane is STRUCTURAL, like Extrude's profile source. While EDITING it is preserved
        // from the seeded original — the card carries no plane control and the old combo silently
        // collapsed a face plane to a base plane through the modeling origin. While ADDING it
-        // comes from what is picked in the viewport. snaporca-e1p.
+        // comes from what is picked in the viewport. e1p.
        if (!editing) { wxString where; f.plane = sketch_plane_from_selection(where); }
        f.width  = m_width->GetValue();
        f.height = m_height->GetValue();
@@ -10032,7 +10032,7 @@ void DesignPanel::update_fillet_gizmo()
 }

 // Push the active Hole card's plane + position + diameter/depth to the viewport gizmo.
-// Grey the FEATURE buttons whose tool cannot run yet, and say why in the tooltip (snaporca-o9j).
+// Grey the FEATURE buttons whose tool cannot run yet, and say why in the tooltip (o9j).
 // Tommaso reported the array controls as MISSING; they were not, but Pattern with no body
 // accepted the click, opened nothing, and wrote its refusal somewhere other than where the click
 // happened — from the user's seat that is indistinguishable from a dead button. A control that
@@ -57,7 +57,7 @@ public:
    void clear_document();      // New Project / Open Project: drop the document with the project
    // Rebuild off the UI thread (progress dialog only if it turns out to be slow), so a feature
    // op on a heavy imported solid does not freeze the window. Returns m_doc.recompute()'s result.
-    // Push the document's recipe into the Model so ANY save path persists it (snaporca-vjk5).
+    // Push the document's recipe into the Model so ANY save path persists it (vjk5).
    void sync_recipe_to_model();
    bool recompute_guarded(const wxString& message);

@@ -173,7 +173,7 @@ private:
    // Which body a tool should act on when it opens: the one picked in the VIEWPORT, else
    // the first. Selection comes first and the tool consumes it — every body combo used to
    // default to index 0, so picking body 3 and opening Mirror silently mirrored body 1.
-    // Clamped to the list, so it is safe to hand straight to SetSelection. snaporca-e1p.
+    // Clamped to the list, so it is safe to hand straight to SetSelection. e1p.
    int  selected_body_default() const;
    void populate_body_choices(int as_of_feature = -1);
    // Fill `c` with the bodies as they existed just before `as_of_feature` and select
@@ -279,7 +279,7 @@ private:
    // The plane the Thread tool builds on: a picked cylindrical face (axis) or the dropdown.
    SketchPlane thread_plane() const;
    // Name the geometry the card has LATCHED, so it never has to be inferred from the viewport.
-    // Pass -1 for "none, falling back to the plane dropdown". See snaporca-200.
+    // Pass -1 for "none, falling back to the plane dropdown". See 200.
    void        set_hole_target_label(int face);
    void        set_thread_target_label(int face, int edge);
    CadFeature build_candidate(Tool t) const;
@@ -489,13 +489,13 @@ private:
    // Polygon's two parameters are chosen FROM THE TOOL, in the offer's Polygon submenu, not
    // from a card on the left: the side count cannot be edited after drawing (the inline editor
    // offers Side and Angle only), so it has to be settled at the moment the tool is armed —
-    // which is exactly where the offer already is. snaporca-e1p.
+    // which is exactly where the offer already is. e1p.
    int               m_poly_sides{6};           // 3..64; the submenu names the common ones
    bool              m_poly_circumscribed{false};

    // Which reference plane a sketch falls back to when no face is picked: 0/1/2 = XY/XZ/YZ,
    // >=3 indexes resolve_datum_planes(). Set by CLICKING a ghost plane in the viewport — there is
-    // deliberately no dropdown for it. snaporca-e1p.
+    // deliberately no dropdown for it. e1p.
    int               m_ref_plane{0};
    // m_ref_plane is always a VALID plane, so it cannot itself distinguish "the user chose XY"
    // from "nobody has chosen anything yet". This does.
@@ -703,7 +703,7 @@ private:
    // The face actually under the last solid click, INDEPENDENT of the whole/face/edge cycle level.
    // The first click on a solid selects the WHOLE body, but the ray has already resolved which face
    // it hit and the callback passes it. "Sketch on the face I clicked" must not require discovering
-    // that a second click refines the selection, so keep it instead of throwing it away. snaporca-3a2.
+    // that a second click refines the selection, so keep it instead of throwing it away. 3a2.
    int               m_pick_face_body{-1};
    int               m_pick_face{-1};
    // What the live sketch was actually opened on ("the picked face", "XY", a datum's name), so the
@@ -767,7 +767,7 @@ private:
    double            m_hole_umin{0}, m_hole_umax{0}, m_hole_vmin{0}, m_hole_vmax{0};
    // Says which face the latch above is holding. Thicken/Shell/Draft show theirs because their
    // face IS the live selection; this one has to be shown precisely BECAUSE it is not, and the
-    // status line goes on saying "Nothing selected" while the ghost keeps drilling. snaporca-200.
+    // status line goes on saying "Nothing selected" while the ghost keeps drilling. 200.
    wxStaticText*     m_hole_target_label{nullptr};

    ComboBox*         m_thread_plane{nullptr};
@@ -885,7 +885,7 @@ private:
    // The guidance sentence for the step the armed sketch tool is on, kept so a transient
    // readout (the live length/angle while a segment is being dragged) can be appended to it
    // instead of replacing it — the guidance used to vanish on the first mouse move after a
-    // click, which is precisely when it is needed. snaporca-1c0c.
+    // click, which is precisely when it is needed. 1c0c.
    wxString          m_sketch_step;
    // mode is a DesignSketchTool::Mode; passed as an int because this header deliberately does
    // not include the tool's, and the .cpp (which does) casts it back.
@@ -308,7 +308,7 @@ void DesignSketchTool::set_tool(Mode mode)

 void DesignSketchTool::cancel()
 {
-    close_session_chrome();     // same orphaned-field freeze as finish() — see snaporca-yce
+    close_session_chrome();     // same orphaned-field freeze as finish() — see yce
    m_active = false;
    m_step_mode_last = -1;
    m_points.clear();
@@ -453,7 +453,7 @@ void DesignSketchTool::delete_selected()
    // now-deleted entity and freeze the flow" — it was simply never called from here. Measured:
    // delete a rectangle whose Width/Height were still queued, draw a circle, type its radius —
    // the field opens, the digits go in, and the radius does not move, because the field belongs
-    // to a rectangle that no longer exists. snaporca-ua9g.
+    // to a rectangle that no longer exists. ua9g.
    reset_autoedit();

    // And re-solve, so the sketch's reported degrees of freedom describe the sketch that is
@@ -463,7 +463,7 @@ void DesignSketchTool::delete_selected()
    if (on_selection_changed) on_selection_changed(0);
 }

-// Convert the selection to/from construction geometry (snaporca-6zic). The Construction
+// Convert the selection to/from construction geometry (6zic). The Construction
 // checkbox only ever set the mode for what you draw NEXT, so a line drawn as real geometry
 // could never become a guide, nor a guide become real. Whole Feature groups flip together:
 // a rectangle is four Line entities and converting three of them is never what was meant.
@@ -1390,7 +1390,7 @@ std::string DesignSketchTool::dimtype_title(DimType k) const {
 // tool at once, which reads like a total product failure and is not necessarily one.
 static void trace_autoedit(const char* why, size_t n)
 {
-    if (!std::getenv("SNAPORCA_UXTRACE")) return;
+    if (!std::getenv("ORCA_CAD_UXTRACE")) return;
    fprintf(stderr, "[UX] autoedit %s steps=%zu\n", why, n);
    fflush(stderr);
 }
@@ -2408,7 +2408,7 @@ bool DesignSketchTool::try_add_constraints(const std::vector<SketchEntityConstra
    m_constraints.resize(mark);                 // roll back the conflicting batch
    // No re-solve to "restore": a failed solve no longer touches the geometry
    // (SketchSolver.cpp only writes back on success), so m_entities still holds the
-    // prior solved state exactly. snaporca-pl5.
+    // prior solved state exactly. pl5.
    return false;
 }

@@ -2516,11 +2516,11 @@ void DesignSketchTool::infer_auto_constraints(int base, double ang_tol_rad, doub
    //    never costs the others. Every rule only pins a relation that is ALREADY true, so
    //    nothing the user drew is moved by this.
    //    A SCRIPTED ADD IS NOT A DRAWN GESTURE — the rule this function already states at its
-    //    bulk call site, which passes zero tolerances for exactly that reason (snaporca-8xg1).
+    //    bulk call site, which passes zero tolerances for exactly that reason (8xg1).
    //    Relational inference must obey it too, and for a second reason beyond tolerance:
    //    EqualRadius couples entities that are geometrically far apart, so on a real drawing it
    //    merges independent connected components into one huge system and defeats the
-    //    component partitioning that makes large sketches solvable at all (snaporca-yww4).
+    //    component partitioning that makes large sketches solvable at all (yww4).
    //    Measured 2026-08-31 on the corpus rung: geometry stayed correct (32/32 sheets clean)
    //    but seven of the largest sheets hit main-thread timeout — MPD681 among them, the very
    //    sheet named in the comment at the bulk call site. Exact-equality would not save it
@@ -2651,7 +2651,7 @@ bool DesignSketchTool::add_imported_regions(
    // Art is not "just drawn", so it must NOT enter the draw-then-edit queue. Without this the
    // glyph contours are treated as fresh entities and a Length field opens on the first of
    // them — on a word, that is one value editor per segment, and an open field freezes the
-    // canvas (snaporca-yce). reset_autoedit() marks every entity as already seen.
+    // canvas (yce). reset_autoedit() marks every entity as already seen.
    reset_autoedit();
    // The new lines carry no constraints, so the solver has nothing to move; resolve anyway so
    // the degrees-of-freedom readout counts them instead of going stale.
@@ -3122,7 +3122,7 @@ void DesignSketchTool::hit_display_sketch(const DisplaySketch& d, const Vec2d& p
 {
    const std::vector<RegionLoop> loops = region_loops(d.entities);
    // What did the sketch decompose into, and what is under the click? This is the trace that
-    // settled snaporca-txp8 — it prints the loop table with each loop's hole count, so
+    // settled txp8 — it prints the loop table with each loop's hole count, so
    // "containment is wrong" and "the click landed elsewhere" stop being indistinguishable.
    // Guarded rather than merely silent: hit_display_sketch runs on every pick, and the message
    // costs a string build and a heap allocation per loop even when nothing consumes it.
@@ -3160,7 +3160,7 @@ void DesignSketchTool::hit_display_sketch(const DisplaySketch& d, const Vec2d& p
            if (h >= 0 && h < int(loops.size()) && point_in_poly(p, loops[h].poly)) { in_hole = true; break; }
        if (!in_hole) { face_feat = d.feature; face_reg = r; }
    }
-    // edge_ent is printed because it is now DELIVERED (snaporca-3648) — a tool can ask for the
+    // edge_ent is printed because it is now DELIVERED (3648) — a tool can ask for the
    // line you pointed at, not just its loop, and "which entity did that click resolve to" is
    // otherwise unanswerable from outside.
    dp_pick_trace("region hit -> feat=%d reg=%d (edge_feat=%d edge_reg=%d edge_ent=%d)",
@@ -3264,11 +3264,11 @@ void DesignSketchTool::select_body(int body)

 // Pick tracing. Selection failures on a real desktop have repeatedly turned out to be an
 // event that never arrived rather than a ray that missed, and the two look identical from
-// the UI. Set SNAPORCA_PICK_TRACE=1 and the whole press->release->ray path narrates itself
+// the UI. Set ORCA_CAD_PICK_TRACE=1 and the whole press->release->ray path narrates itself
 // on stderr. Off by default: no cost, no noise.
 static bool dp_pick_trace_on()
 {
-    static const bool on = ::getenv("SNAPORCA_PICK_TRACE") != nullptr;
+    static const bool on = ::getenv("ORCA_CAD_PICK_TRACE") != nullptr;
    return on;
 }

@@ -3294,7 +3294,7 @@ static void dp_pick_trace(const char* fmt, ...)
 //
 // ponytail: crossing over a triangle sample set. A rectangle small enough to sit entirely
 // inside one flat triangle selects nothing — drag a bigger one, or click. Real multi-body
-// selection (and the homogeneous-set rule that goes with it) is snaporca-9xw.
+// selection (and the homogeneous-set rule that goes with it) is 9xw.
 void DesignSketchTool::pick_bodies_in_rectangle()
 {
    if (m_solid_mesh == nullptr || m_solid_tri_body == nullptr || m_solid_bodies == nullptr)
@@ -3486,7 +3486,7 @@ bool DesignSketchTool::handle_solid_click(GLCanvas3D& canvas, const wxMouseEvent
    m_sel_vertex_pt = p.vertex_pt;
    m_solid_sel     = p.kind;

-    // CLICK AGAIN ON THE SAME THING -> THE WHOLE BODY (snaporca-gem). Pointing at a face and
+    // CLICK AGAIN ON THE SAME THING -> THE WHOLE BODY (gem). Pointing at a face and
    // pointing at its body are different intents, and until now only the rubber band could
    // express the second one — so the status line said "face 0 selected" while the user
    // believed they had taken the body, and every body verb had to opt into the face kinds to
@@ -3823,23 +3823,23 @@ void DesignSketchTool::clear_extrude_gizmo()
 // cone travels, an open collar receives. No surveyed CAD system encodes this at all; both ends of
 // their mates are drawn identically, which is why "which part moves?" is a standing complaint.
 //
-// SNAPORCA_GLYPH=A|B selects the treatment while this is being judged on the rig:
+// ORCA_CAD_GLYPH=A|B selects the treatment while this is being judged on the rig:
 //   A  three short axis arms, no head differentiation  (the Onshape baseline)
 //   B  one-sided Z arrow, filled vs open head          (the proposal)          -- default
 void DesignSketchTool::render_mate_connectors()
 {
    if (m_mate_connectors.empty()) return;
    static const bool style_A = [] {
-        const char* s = ::getenv("SNAPORCA_GLYPH");
+        const char* s = ::getenv("ORCA_CAD_GLYPH");
        return s && (*s == 'A' || *s == 'a');
    }();
    // The face treatment, on by default. Read every frame rather than latched in a static, so
    // toggling the preference takes effect on the next repaint instead of at the next launch —
    // it is a look, and a look you cannot A/B without restarting will not get compared.
-    // SNAPORCA_GLYPH=D forces the disc regardless, which is how the rig drives the other branch.
+    // ORCA_CAD_GLYPH=D forces the disc regardless, which is how the rig drives the other branch.
    const bool face_style = !style_A
                         && wxGetApp().app_config->get_bool("design_connector_face_glyph")
-                         && [] { const char* s = ::getenv("SNAPORCA_GLYPH");
+                         && [] { const char* s = ::getenv("ORCA_CAD_GLYPH");
                                 return !(s && (*s == 'D' || *s == 'd')); }();

    const Camera& cam = wxGetApp().plater()->get_camera();
@@ -4021,7 +4021,7 @@ void DesignSketchTool::render_mate_connectors()
 }

 // ---------------------------------------------------------------------------------------------
-// THE FACE TREATMENT of the mate connector (snaporca-x0kd). The disc + roll quadrant answers
+// THE FACE TREATMENT of the mate connector (x0kd). The disc + roll quadrant answers
 // "where is X" with a shape that has to be learned; a face does not. Face orientation is
 // hardwired perception -- a toddler reads a face's roll and verse with no instruction at all --
 // and that is the whole reason this exists. Default ON, switchable in Preferences for users who
@@ -4053,7 +4053,7 @@ static const Vec2d kBearOutline[] = {        // 12 verts, RDP eps 0.030, CCW
 static const Vec2d kBearChin[] = {            // the CHIN BAR, flat. The muzzle is relief — see kBearCrest.
    {-0.2682, -0.3578}, {+0.2628, -0.3578}, {+0.2237, -0.1786},
 };
-// {cx, cy, r}: two eyes, then the cheek dot that carries handedness (snaporca-wi3z).
+// {cx, cy, r}: two eyes, then the cheek dot that carries handedness (wi3z).
 static const Vec3d kBearMarks[] = {
    {-0.1997, +0.1760, +0.0590},
    {+0.1947, +0.1760, +0.0590},
@@ -6436,7 +6436,7 @@ DesignSketchTool::region_loops(const std::vector<SketchEntity>& ents) const

    // NESTING. A loop drawn inside another one is that one's HOLE. Without this a sketch is
    // just N disjoint filled polygons, so "the plate with the hole" is not expressible and the
-    // multi-loop kernel path (snaporca-88v) is unreachable from the viewport — which is exactly
+    // multi-loop kernel path (88v) is unreachable from the viewport — which is exactly
    // what Tommaso hit: a rectangle with a circle inside extruded to a plain box, because only
    // the rectangle loop could be picked and only its entities were passed on.
    //
@@ -6450,7 +6450,7 @@ DesignSketchTool::region_loops(const std::vector<SketchEntity>& ents) const
    // polygon being tested answers by rounding, so the same drawing can be read either way.
    // Measured on the StudyCadCam corpus: the engine and an independent containment check
    // disagreed on 6 of 39 sheets, and every disagreement was a probe point sitting on the other
-    // loop's boundary. snaporca-5hvl.
+    // loop's boundary. 5hvl.
    auto poly_area = [](const std::vector<Vec2d>& q) {
        double a2 = 0.0;
        for (size_t i = 0, j = q.size() - 1; i < q.size(); j = i++)
@@ -6550,7 +6550,7 @@ int DesignSketchTool::region_at(const Vec2d& p) const

 // ---- rendering --------------------------------------------------------------

-// Chop a polyline into dashes (snaporca-imlq). Construction geometry is dashed in every CAD;
+// Chop a polyline into dashes (imlq). Construction geometry is dashed in every CAD;
 // this one painted it solid grey, which against the under-constrained orange reads as "another
 // line", not as "reference only". The dash and gap arrive in WORLD units — the caller scales them
 // by units-per-pixel, so the dash keeps its size on screen at any zoom instead of turning into a
@@ -7986,7 +7986,7 @@ void DesignSketchTool::confirm_op()
        // The sources as they stand BEFORE any of this op's constraints exist. Two jobs: every
        // copy is reflected from the untouched original (so a batch that moves the sketch cannot
        // feed a later copy moved geometry), and the invariant at the bottom has something to
-        // compare against. snaporca-mirror-slot.
+        // compare against. mirror-slot.
        const std::vector<SketchEntity> before = m_entities;
        const size_t cmark = m_constraints.size();
        std::vector<std::pair<int, SketchEntity>> fresh;   // copy index -> its pristine reflection
@@ -8033,7 +8033,7 @@ void DesignSketchTool::confirm_op()
        // postcondition on the geometry, and if a source moved it keeps the copies — which are
        // exactly what the preview showed — and drops the whole constraint web that moved them.
        // Restoring the sources needs no re-solve: the pre-batch state was itself solved, and a
-        // failed solve does not write back (snaporca-pl5).
+        // failed solve does not write back (pl5).
        // BOTH HALVES. Watching only the sources caught the slot (whose web dragged everything)
        // and missed the rounded rectangle, where the solver held the sources still and put the
        // COPIES somewhere else: an arc has five degrees of freedom and Symmetric on centre plus
@@ -8449,7 +8449,7 @@ const ColorRGBA* DesignSketchTool::sketch_hl_color(int feature) const
    return nullptr;
 }

-// Which step of the armed gesture is live, reported only when it moves (snaporca-1c0c). Called
+// Which step of the armed gesture is live, reported only when it moves (1c0c). Called
 // from render(), which is the one place EVERY state change passes through — a per-call-site
 // notification would have to be added to each of the thirty-odd tool branches and would be
 // forgotten by the next one. Cheap: three ints compared per frame.
@@ -8753,7 +8753,7 @@ void DesignSketchTool::render(GLCanvas3D& canvas)
        // Mirror's is the axis, Fillet/Chamfer's is the first of the two lines — and until now
        // every pick painted the same white, so the picture could not answer "what did I select
        // as what". Violet, not cyan: cyan means SELECTED here and nothing else may wear it.
-        // snaporca-vd6v.
+        // vd6v.
        const bool op_ref = is_edit_op_mode() && int(i) == m_op_a;
        ColorRGBA col;
        if (editing_this)        col = editing;
@@ -8833,7 +8833,7 @@ void DesignSketchTool::render(GLCanvas3D& canvas)
        if (!sel_handles.empty()) draw_vertices(m_highlight_model, sel_handles, sel_col);

        // Midpoint of every segment, drawn smaller and cooler than the endpoint handles
-        // (snaporca-te8v). Without it the Midpoint snap is invisible: it exists in the
+        // (te8v). Without it the Midpoint snap is invisible: it exists in the
        // inference engine but the user has nothing to aim at. Construction lines get one
        // too — you constrain to them as readily as to real geometry.
        std::vector<Vec2d> mids;
@@ -9355,7 +9355,7 @@ int DesignSketchTool::add_entities_scripted(const std::vector<SketchEntity>& ent
    // the 39 corpus drawings the loops that came back wrong were all TINY (1.4 to 13 mm^2), out
    // by up to 7e-4 relative, because a 0.005 degree tilt on a 0.3 mm chord is inside 1e-4.
    // With zero, only a segment that is EXACTLY axis-aligned is constrained, and constraining
-    // something already true cannot move it. snaporca-8xg1.
+    // something already true cannot move it. 8xg1.
    // The weld window closes too. Two endpoints a micron apart are not the same point when a
    // caller typed both of them: on MPD681, 20 of 363 scripted segments were dragged onto a
    // common point up to 0.0021 mm away, because welding is TRANSITIVE and three vertices near
@@ -9370,7 +9370,7 @@ int DesignSketchTool::add_entities_scripted(const std::vector<SketchEntity>& ent
    // while m_awaiting_length) and swallows every letter (in_text includes inline_busy()). The
    // symptom was that the first key and click after sketch_add did nothing until one Escape had
    // dismissed the field. Resyncing the baseline here leaves an ALREADY open field alone; it
-    // only stops this add from being read as something the user just drew. snaporca-j7gc.
+    // only stops this add from being read as something the user just drew. j7gc.
    m_autoedit_seen = int(m_entities.size());
    return base;
 }
@@ -9582,7 +9582,7 @@ bool DesignSketchTool::select_at_screen(GLCanvas3D& canvas, int sx, int sy)
        // counts only m_selection, so right-clicking a sketch point produced the EMPTY
        // vocabulary and every SkPoint row in the atlas was unreachable from the menu. Other
        // entities keep the handle pick: a line's endpoint is a drag target, not a thing with a
-        // vocabulary of its own. snaporca-lnri.
+        // vocabulary of its own. lnri.
        if (ei >= 0 && ei < int(m_entities.size())
            && m_entities[ei].type == SketchEntity::Type::Point) {
            if (std::find(m_selection.begin(), m_selection.end(), ei) != m_selection.end())
@@ -9664,8 +9664,8 @@ std::vector<int> DesignSketchTool::connected_loop(int seed) const
 // suppresses the menu whenever it is set, so right-click became a no-op that also hid the one door
 // to half the vocabulary (47 of 86 verbs have no shortcut). Measured on the rig: with Line armed,
 // two right-clicks in a row produced no menu and no tool change; only Escape freed it.
-// Same rule as snaporca-xmh6, which said it for the selection: clearing nothing is not a gesture
-// terminator. snaporca-ghcz.
+// Same rule as xmh6, which said it for the selection: clearing nothing is not a gesture
+// terminator. ghcz.
 bool DesignSketchTool::right_abandon()
 {
    if (m_points.empty())
@@ -10000,7 +10000,7 @@ bool DesignSketchTool::on_mouse_impl(wxMouseEvent& evt, GLCanvas3D& canvas)
        // consumed from here on. Left-drag no longer orbits in this canvas — DesignCanvas puts
        // orbit on middle-drag and pan on right-drag, the CAD convention — so nothing downstream
        // is being starved of a gesture it used to own.
-        // HOVER PRE-HIGHLIGHT (snaporca-9xw part 3): say what a click would take, before it is
+        // HOVER PRE-HIGHLIGHT (9xw part 3): say what a click would take, before it is
        // taken. Plain motion only — no button down, no band running — because during a drag the
        // pointer is doing something else and a promise about clicking would be a lie. Returns
        // false so the event still reaches the camera; this only asks for a repaint, it does not
@@ -10090,7 +10090,7 @@ bool DesignSketchTool::on_mouse_impl(wxMouseEvent& evt, GLCanvas3D& canvas)
            return true;
        }
        m_display_pick = -1; m_display_pick_region = -1;  // clicked bare plate -> drop highlight
-        // ...and the SOLID selection goes with it (snaporca-od0). A click that hits nothing has to
+        // ...and the SOLID selection goes with it (od0). A click that hits nothing has to
        // mean what a rubber band that sweeps nothing already means — pick_bodies_in_rectangle
        // clears on an empty sweep, and the two gestures cannot disagree about the same outcome.
        // Until now the face survived a click on bare plate, so "click away, then click the face
@@ -10738,7 +10738,7 @@ bool DesignSketchTool::on_mouse_impl(wxMouseEvent& evt, GLCanvas3D& canvas)
            return true;
        }
        if (evt.RightDown() && m_points.empty())
-            return false;               // no chain to end — snaporca-ghcz, let the offer open
+            return false;               // no chain to end — ghcz, let the offer open
        if (evt.RightDown()) {
            // END the chain — do NOT close it. This used to call push_closed_lines() for three
            // or more points, i.e. it drew a final segment from the last point back to the
@@ -11117,7 +11117,7 @@ bool DesignSketchTool::on_mouse_impl(wxMouseEvent& evt, GLCanvas3D& canvas)
            return true;
        }
        if (evt.RightDown() && m_points.empty())
-            return false;               // no poles down — snaporca-ghcz, let the offer open
+            return false;               // no poles down — ghcz, let the offer open
        if (evt.LeftDClick() || evt.RightDown()) {
            if (m_points.size() >= 2) {
                const int base = int(m_entities.size());
@@ -64,7 +64,7 @@ public:
                      Constrain };
    // Which tool is armed, and how many anchors it has down. Read-only, for the offer ladder:
    // "the menu armed the verb I chose" is otherwise unassertable, and a menu walk that lands one
-    // row off arms a NEIGHBOURING tool and then grades whatever that drew. snaporca-ekt9.
+    // row off arms a NEIGHBOURING tool and then grades whatever that drew. ekt9.
    Mode mode() const { return m_mode; }
    int  pending_points() const { return int(m_points.size()); }
    void emit_step_hint();   // fires on_step_changed when the step actually moved
@@ -133,7 +133,7 @@ public:
    bool has_entities() const { return !m_entities.empty(); }
    bool on_mouse(wxMouseEvent& evt, GLCanvas3D& canvas);
    // Right-click on a draw tool: true when an in-progress anchor was abandoned, false when
-    // there was nothing to abandon — and false is what lets the offer menu open. snaporca-ghcz.
+    // there was nothing to abandon — and false is what lets the offer menu open. ghcz.
    bool right_abandon();
    // True if the LAST right-press was consumed as a gesture terminator (end a polyline chain,
    // abandon an anchor, exit a tool). Read-and-clear: the canvas asks on the matching release to
@@ -213,7 +213,7 @@ public:
    // feature index + the clicked closed-region index within it (-1 = no specific loop).
    // entity = the sketch entity index under the cursor when the click landed on a loop
    // STROKE, else -1 for an interior/region hit. Carried because a tool can legitimately
-    // want the LINE you pointed at, not just the loop it belongs to (Rib, snaporca-3648).
+    // want the LINE you pointed at, not just the loop it belongs to (Rib, 3648).
    std::function<void(int feature, int region, int entity)> on_display_sketch_selected;
    // Double-click on a committed sketch stroke: open THAT feature for editing. Selecting a line
    // and then hunting for an Edit button in a panel is the dependency this tab exists to remove.
@@ -342,7 +342,7 @@ public:

    // Mate connectors. Until now a connector was visible only to a program — resolve_datum_coordsys
    // had exactly one consumer, the MCP socket — so the frame a mate is built on could not be seen
-    // at all. The glyph has to answer two questions on sight (snaporca-wgsc): which way does Z point
+    // at all. The glyph has to answer two questions on sight (wgsc): which way does Z point
    // (the VERSE), and which of the pair is anchored versus driven (the POLARITY). Nothing in any
    // surveyed CAD system encodes the second one.
    struct MateConnectorGlyph {
@@ -440,7 +440,7 @@ public:
    // Live readout while drawing a Line/Polyline segment (anchor->cursor metrics).
    std::function<void(double length, double angle_deg, bool locked)> on_cursor_metrics;

-    // Live step guidance (snaporca-1c0c). The armed tool reports WHICH STEP of its gesture the
+    // Live step guidance (1c0c). The armed tool reports WHICH STEP of its gesture the
    // user is on, every time that changes, so the status line can name the next click instead of
    // repeating the one-shot sentence written when the tool was armed. step = anchors/picks
    // already down (Mirror: 0 = no axis, 1 = axis down, 2 = ready to apply); picks = size of the
@@ -679,7 +679,7 @@ private:
    // weld_tol:    how far apart two endpoints may be and still be called Coincident.
    // Both default to GESTURE slack. A scripted add passes zero for both: the caller has
    // already said exactly what it means, and every non-zero window is a window in which the
-    // inference rewrites it. snaporca-8xg1.
+    // inference rewrites it. 8xg1.
    void infer_auto_constraints(int base, double ang_tol_rad = 3.0 * M_PI / 180.0,
                                double weld_tol = 1e-3);

@@ -939,7 +939,7 @@ private:
    // A selectable sketch region: its own boundary, plus the loops nested INSIDE it, which
    // are its holes. Modelling holes is what makes "the plate with the hole in it" a thing the
    // user can point at — without it a sketch is N disjoint filled polygons and the only
-    // selectable things are the rectangle alone or the circle alone (snaporca-txp8).
+    // selectable things are the rectangle alone or the circle alone (txp8).
    struct RegionLoop {
        std::vector<Vec2d> poly;
        std::vector<int>   ents;
@@ -1165,7 +1165,7 @@ private:
        Vec3d              vertex_pt{Vec3d::Zero()};
    };
    bool resolve_solid_pick(GLCanvas3D& canvas, int mx, int my, SolidPick& out) const;
-    // HOVER PRE-HIGHLIGHT (snaporca-9xw part 3). Vertex-beats-edge-beats-face is a rule the user
+    // HOVER PRE-HIGHLIGHT (9xw part 3). Vertex-beats-edge-beats-face is a rule the user
    // cannot see until after they commit to a click; showing the outcome under the pointer is
    // what makes the precedence learnable at all, and is the charter's L5 (one click, one visible
    // change) read honestly — the change has to be predictable before the click, not only after.
@@ -113,7 +113,7 @@ json describe_tools()
    // Hand-written descriptor. The bridge turns this into MCP tool schemas; later
    // slices grow this list (ideally from the kernel directly).
    return json{
-        {"app", "SnapOrca CAD"},
+        {"app", "Orca CAD"},
        {"protocol", "jsonrpc-2.0"},
        {"slice", 5},
        // Read this before using any face or edge id.
@@ -2201,9 +2201,9 @@ void server_thread(std::string sock_path)

 void start_mcp_control_if_enabled()
 {
-    const char* env = std::getenv("SNAPORCA_MCP");
+    const char* env = std::getenv("ORCA_CAD_MCP");
    if (!env || !*env) return;
-    std::string path = (std::strcmp(env, "1") == 0) ? "/tmp/snaporca-mcp.sock" : env;
+    std::string path = (std::strcmp(env, "1") == 0) ? "/tmp/orca-cad-mcp.sock" : env;
    static bool started = false;
    if (started) return;
    started = true;
@@ -3,9 +3,9 @@

 // MCP control surface (slice 1): a local JSON-RPC 2.0 server, line-delimited over a
 // Unix domain socket, that lets an external MCP bridge drive and perceive the Design
-// tab. Off unless the env var SNAPORCA_MCP is set:
-//   SNAPORCA_MCP=1                -> socket at /tmp/snaporca-mcp.sock
-//   SNAPORCA_MCP=/path/to.sock    -> socket at that path
+// tab. Off unless the env var ORCA_CAD_MCP is set:
+//   ORCA_CAD_MCP=1                -> socket at /tmp/orca-cad-mcp.sock
+//   ORCA_CAD_MCP=/path/to.sock    -> socket at that path
 // All CAD work is marshalled onto the wx main thread and runs through the SAME
 // CadDocument kernel the GUI uses (no parallel engine). Slice-1 methods:
 //   describe_tools, describe_scene, extrude.
@@ -15,7 +15,7 @@

 namespace Slic3r { namespace GUI {

-// Start the server thread iff SNAPORCA_MCP is set. Safe to call once after the
+// Start the server thread iff ORCA_CAD_MCP is set. Safe to call once after the
 // MainFrame + DesignPanel exist. No-op when the env var is unset or on Windows.
 void start_mcp_control_if_enabled();

@@ -52,7 +52,7 @@ bool parse_value(const char* text, double& out)

 // One machine-readable line per event of the click-edit contract, for the UX check that runs
 // after every build (scripts/CAD/check-gui-click-edit.py). Deliberately NOT the same switch as
-// SNAPORCA_KEYTRACE: that one is a debugging firehose, this one is an assertion surface and its
+// ORCA_CAD_KEYTRACE: that one is a debugging firehose, this one is an assertion surface and its
 // format is a contract the script parses.
 //
 // The pair that matters is `open` vs `commit`: the check always types a value DIFFERENT from the
@@ -61,7 +61,7 @@ bool parse_value(const char* text, double& out)
 // the user actually gets can.
 void ux_trace(const char* event, const std::string& title, const std::string& detail)
 {
-    if (!std::getenv("SNAPORCA_UXTRACE")) return;
+    if (!std::getenv("ORCA_CAD_UXTRACE")) return;
    std::fprintf(stderr, "[UX] %s title=%s %s\n", event, title.c_str(), detail.c_str());
    std::fflush(stderr);
 }
@@ -175,7 +175,7 @@ bool SketchInlineEditor::render(ImGuiWrapper& imgui, float scale)
    // different causes — no FRAMES (this canvas repaints on demand only, so an idle canvas never
    // processes ImGui's queued characters) versus frames that run while the input is not active —
    // and they are indistinguishable from outside.
-    if (std::getenv("SNAPORCA_UXTRACE")) {
+    if (std::getenv("ORCA_CAD_UXTRACE")) {
        const ImGuiIO& io = ImGui::GetIO();
        std::fprintf(stderr, "[UX] frame title=%s want_text=%d want_kb=%d active=%d buf=%s\n",
                     m_title.c_str(), (int) io.WantTextInput, (int) io.WantCaptureKeyboard,
@@ -517,7 +517,7 @@ bool ImGuiWrapper::update_key_data(wxKeyEvent &evt)
        // runs handlers in reverse bind order, and on_char returns without Skip() whenever this
        // function returns true — so such a probe stays silent whether or not the key arrived.
        // A day was lost to reading that silence as evidence.
-        if (std::getenv("SNAPORCA_UXTRACE")) {
+        if (std::getenv("ORCA_CAD_UXTRACE")) {
            fprintf(stderr, "[UX] imgui_char unicode=%d keycode=%d want_text=%d\n",
                    (int) key, evt.GetKeyCode(), (int) io.WantTextInput);
            fflush(stderr);
@@ -1361,7 +1361,7 @@ void MainFrame::init_tabpanel() {
        m_design_page = new wxPanel(this);
        m_design_page->SetSizer(new wxBoxSizer(wxVERTICAL));
        m_design_page->Hide();
-        start_mcp_control_if_enabled();   // opens the MCP socket iff SNAPORCA_MCP is set
+        start_mcp_control_if_enabled();   // opens the MCP socket iff ORCA_CAD_MCP is set
    }
 #endif

@@ -1,7 +1,7 @@
 #include <catch2/catch_all.hpp>   // mainline OrcaSlicer ships Catch2 v3 (v2 was catch2/catch.hpp)

 // Substring assertions, spelled so this file compiles UNCHANGED on both forks.
-// Catch2 v2 (snaporca) spells it Matchers::Contains; v3 (orca_cad / mainline) spells it
+// Catch2 v2 (Snapmaker) spells it Matchers::Contains; v3 (orca_cad / mainline) spells it
 // Matchers::ContainsSubstring and gives Contains an incompatible meaning — range-contains-
 // ELEMENT — which fails to compile against a std::string rather than failing a test.
 // Using find() sidesteps the rename entirely; INFO keeps the actual string in the report.
@@ -613,7 +613,7 @@ TEST_CASE("entity constraints: point-on-line positions a centre onto an axis", "
 // vendored solver (slvs/dsc.h FindById, "Cannot find handle"), taking every later test with
 // it, and was quarantined for it. Fixed in SketchSolver: a full circle can no longer be handed
 // to SLVS_C_ARC_LINE_TANGENT, which dereferences arc endpoints a circle does not have. See
-// snaporca-tkz.
+// tkz.
 TEST_CASE("entity constraints: tangent/midpoint/symmetric/angle", "[CadDocument][sketch]")
 {
    using R = SketchPointRole;
@@ -962,7 +962,7 @@ TEST_CASE("extrude taper + up-to-face distance", "[CadDocument]")
    }
 }

-// Was [known-broken] until the numbers were actually measured (snaporca-kzy). The geometry
+// Was [known-broken] until the numbers were actually measured (kzy). The geometry
 // was right all along; the TEST compared against the wrong reference. An internal thread bores
 // at the MINOR radius (radius - depth) and then carves the groove out to radius + depth, so a
 // tapped hole keeps the crests between turns and therefore holds MORE material than a plain
@@ -1330,7 +1330,7 @@ TEST_CASE("datum plane: offset + tilt resolution and sketching on it", "[CadDocu

    // A datum-plane-only document has no solid, and that is a benign SUCCESS, not a benign
    // failure. It used to return false, and "benign failure" is exactly the phrasing that hid
-    // snaporca-mtav: two callers read the false as "unusable document" and threw the design
+    // mtav: two callers read the false as "unusable document" and threw the design
    // away — the 3MF recipe was never written, and a project that had one was refused on load.
    CadDocument only_plane;
    only_plane.add_plane(0, 10.0, 0.0, 0, "P");
@@ -1416,7 +1416,7 @@ TEST_CASE("draft tapers a solid face about the body base", "[CadDocument]")

 TEST_CASE("a split renumbers the bodies a later feature indexes", "[CadDocument][cut]")
 {
-    // Pins the invariant the Design tab's re-edit path depends on (snaporca-oz7): a stored
+    // Pins the invariant the Design tab's re-edit path depends on (oz7): a stored
    // target_body indexes the body list AS IT WAS when that feature ran, and a Cut placed
    // later in the tree changes that list. If this test ever fails, the GUI's
    // fill_body_choice() replay-to-timeline-slot assumption needs revisiting with it.
@@ -2502,7 +2502,7 @@ TEST_CASE("datum plane construction methods", "[CadDocument][plane]")
    }
 }

-// Mirrors the snaporca [Deviation] case (snaporca carries it in test_geometry.cpp; here it lives
+// Mirrors the Snapmaker [Deviation] case (Snapmaker carries it in test_geometry.cpp; here it lives
 // alongside the CAD suite). GeometryEngine::surface_deviation = one-sided Hausdorff used by the
 // MCP validate_against acceptance metric.
 TEST_CASE("surface_deviation: identical solids ~0, shifted solid ~shift", "[Deviation]")
@@ -5310,7 +5310,7 @@ TEST_CASE("thicken-surface makes a solid from a sheet", "[CadDocument][surface]"
    REQUIRE_THAT(double(tx1), WithinAbs(double(sx1), 2.1));
 }

-// Regression for snaporca-lu27, with the rig's own numbers. A 60x60x40 four-walled open box
+// Regression for lu27, with the rig's own numbers. A 60x60x40 four-walled open box
 // used to report volume 96000 and an inertia diagonal of [-4.2e7, -4.2e7, -6.9e7] — negative
 // principal moments, which no real body can have. VolumeProperties was being integrated over
 // an open shell as though it were closed, and std::abs() on the mass hid the only obvious tell.
@@ -5352,7 +5352,7 @@ TEST_CASE("mass properties of a sheet body report area only, never a volume",
    REQUIRE(solid.inertia[8] > 0.0);
 }

-// snaporca-wm4s. The wall of a thickened open box must contain the corner material. Thickening
+// wm4s. The wall of a thickened open box must contain the corner material. Thickening
 // each face along its own normal and sewing (MakeThickSolidBySimple) leaves the four vertical
 // corners empty and measured 29648.15 where the geometry requires 44000; the two controls below
 // were exact before and must stay exact, since they are what a corner-only fix must not disturb.
@@ -7309,7 +7309,7 @@ TEST_CASE("interference: detects a clash created by a mate", "[CadDocument][inte
 // A filleted solid must reach the plate as a watertight mesh. OCCT emits one degenerate
 // triangle at the pole of every corner sphere patch; welded, its v->v edge counts as an open
 // edge and the slicer tells the user to go repair the model in another CAD application --
-// the exact round trip this feature exists to remove. snaporca-agw.
+// the exact round trip this feature exists to remove. agw.
 TEST_CASE("CadDocument filleted solid tessellates watertight", "[CadDocument]")
 {
    CadDocument doc;
@@ -7339,7 +7339,7 @@ TEST_CASE("CadDocument filleted solid tessellates watertight", "[CadDocument]")
 // into the params whether or not it converged, so reading geometry back unconditionally made
 // every failed attempt destructive -- and the fillet degrade ladder tries a deliberately
 // over-constrained rung FIRST, so a filleted corner was wrecked before the rung that works
-// ever got a chance. snaporca-pl5.
+// ever got a chance. pl5.
 TEST_CASE("Failed sketch solve leaves geometry untouched", "[CadDocument]")
 {
    using R  = SketchPointRole;
@@ -7415,7 +7415,7 @@ TEST_CASE("Failed sketch solve leaves geometry untouched", "[CadDocument]")
 // A subtraction whose tool misses the target is a perfectly legal boolean that removes nothing,
 // so OCCT reports success and the feature lands in the recipe with ok:true and an unchanged body.
 // That is how a hole placed with world coordinates instead of plane-frame ones read as "drilled"
-// three times in a row while the volume never moved. snaporca-daf.
+// three times in a row while the volume never moved. daf.
 TEST_CASE("A cut that removes no material is an error, not a silent success", "[CadDocument]")
 {
    // 20 x 20 box, 20 tall, centred on the origin of the XY plane.
@@ -7472,7 +7472,7 @@ TEST_CASE("A cut that removes no material is an error, not a silent success", "[
 // Anything else returns a null wire, and build_sketch_wire used to answer that by falling through
 // to its legacy tail — which ends in a rectangle built from width/height. For an entity sketch
 // those are whatever they were initialised to, so the extrude produced a box nobody drew.
-// snaporca-88v.
+// 88v.
 TEST_CASE("An entity sketch that forms no wire fails instead of extruding a default box", "[CadDocument]")
 {
    auto circle = [](Vec2d c, double r) {
@@ -7562,7 +7562,7 @@ CadDocument plate_doc(const std::vector<SketchEntity>& entities, double distance

 } // namespace

-// snaporca-88v: a sketch may hold more than one closed loop. The Extrude path builds the
+// 88v: a sketch may hold more than one closed loop. The Extrude path builds the
 // sketch's planar region via SketchEngine::entities_to_wires + wires_to_face: the largest loop
 // is the outer boundary, every other loop a hole. Volumes are the proof — a plate with a hole
 // must subtract the hole, not merely "not throw".
@@ -7637,7 +7637,7 @@ TEST_CASE("entities_to_wires returns one wire per loop", "[CadDocument][sketchwi
 // Sketching on a picked face is the most common gesture in solid modelling, and it was impossible:
 // the plane came from a combo of base + datum planes only, so the sole route onto a face was to
 // build a Coincident datum plane first. plane_of_face is the shared derivation that makes the
-// viewport selection usable directly. snaporca-3a2.
+// viewport selection usable directly. 3a2.
 TEST_CASE("plane_of_face gives a sketchable plane for a planar face only", "[CadDocument]")
 {
    // 20 x 20 x 20 box on XY, so its top face sits at z = 20 with +Z normal.
@@ -7689,7 +7689,7 @@ TEST_CASE("plane_of_face gives a sketchable plane for a planar face only", "[Cad
    }
 }

-// snaporca-5425 — POSITIVE-CONTRACT variant. A feature that left a body with a null
+// 5425 — POSITIVE-CONTRACT variant. A feature that left a body with a null
 // TopoDS_Shape used to be tolerated: recompute() returned true and the document kept
 // advertising the body. The new guard makes that a hard failure. This test asserts the
 // contract the guard preserves on the healthy side: a normal box + fillet document
@@ -7727,7 +7727,7 @@ TEST_CASE("recompute on a healthy box + fillet leaves no body null and no error
 }

 // ============================================================================
-// snaporca-rgbj — does a chamfer chain degenerate from a KERNEL defect, or from
+// rgbj — does a chamfer chain degenerate from a KERNEL defect, or from
 // how the DRIVER captured its edge ids? Experiment, not a fix.
 //
 // CadFeature::dressup_edge is a GLOBAL edge id: an ordinal into
@@ -8242,7 +8242,7 @@ TEST_CASE("add_extrude_entities builds a plate with a bore (clockwise circle)",
 }


-// snaporca-mtav. A document that has only sketches in it is not a broken document, it is the
+// mtav. A document that has only sketches in it is not a broken document, it is the
 // state every design passes through between drawing a profile and extruding it. recompute()
 // used to call that "no solid-producing features" and return false, and two things downstream
 // read that false as "the document is unusable": the GUI syncs the 3MF recipe only after a
@@ -113,7 +113,7 @@ TEST_CASE("slvs: over-constrained / inconsistent is detected", "[slvs]")
    CHECK_FALSE(res.ok);   // SLVS_RESULT_INCONSISTENT
 }

-// snaporca-yww4. libslvs sizes its System with a compile-time `MAX_UNKNOWNS = 1024`, and the
+// yww4. libslvs sizes its System with a compile-time `MAX_UNKNOWNS = 1024`, and the
 // solver is handed every entity in the sketch at 2 params per point — so a sketch of about 480
 // lines is the last one that fits and the next comes back TOO_MANY_UNKNOWNS. Because
 // try_add_constraints rolls a failed batch back, that turned into: every auto-inferred constraint
@@ -1,20 +1,20 @@
 #!/usr/bin/env python3
-"""Zero-dependency stdio MCP server bridging to the SnapOrca/Orca-CAD control socket.
+"""Zero-dependency stdio MCP server bridging to the Orca/Orca-CAD control socket.

 Speaks MCP (JSON-RPC 2.0 over newline-delimited stdio) to an MCP client (Claude Code),
 and forwards each tool call to the app's Unix-domain control socket (opened by the GUI
-when launched with SNAPORCA_MCP set). The tool list is built *live* from the app's own
+when launched with ORCA_CAD_MCP set). The tool list is built *live* from the app's own
 `describe_tools` reply — introspection drives the schema, so new kernel methods surface
 without touching this file.

-Usage:  snaporca_mcp_bridge.py [SOCKET_PATH]   (default /tmp/snaporca-mcp.sock)
-The app must be running with SNAPORCA_MCP set; if the socket is down, tools/list falls
+Usage:  orca_cad_mcp_bridge.py [SOCKET_PATH]   (default /tmp/orca-cad-mcp.sock)
+The app must be running with ORCA_CAD_MCP set; if the socket is down, tools/list falls
 back to the slice-1 set and tool calls report the connection error (never crash).
 """
 import sys, os, json, socket, itertools

-SOCK_PATH = sys.argv[1] if len(sys.argv) > 1 else "/tmp/snaporca-mcp.sock"
-SERVER_INFO = {"name": "snaporca-cad", "version": "0.1"}
+SOCK_PATH = sys.argv[1] if len(sys.argv) > 1 else "/tmp/orca-cad-mcp.sock"
+SERVER_INFO = {"name": "orca-cad", "version": "0.1"}
 _app_id = itertools.count(1)

 # --- app control-socket round-trip --------------------------------------------