orthographic-registration — Claude Code skill for Blender views
Registers front, side, back and top orthographic references into one Blender coordinate contract. Useful when the front reads fine but side and back silhouettes drift.
npx skills add RobLe3/cc-blender-skill --skill orthographic-registration -a claude-codeWhat orthographic-registration does
The skill fixes a coordinate contract: front defines the X/Z silhouette, side gives the Y/Z depth envelope, top sets X/Y spread, and back only refines the rear shape without rewriting the front. The workflow follows that split. Run scripts/register_views.py over the view images, build orthographic reference planes at a single shared scale, align the centerline and bbox centers before modeling, lock the front X/Z boundary vertices, then add Y depth from the side and top envelopes with modifiers, displacement or vertex groups. All four cameras get validated before export.
One rotation rule trips people up. The front camera looks along Y and the silhouette sits in X/Z, so a 2D rotation inside the front view is a rotation around Y: rotation_euler = (0, angle, 0). Rotating around Z spins those meshes out of screen space.
When views disagree, the front brand read wins and the conflict goes into registration_report.json. You need a Blender MCP server for this to do anything: the skill calls it for code execution, object summaries, viewport renders and API doc search. Before the first Blender call, read ../text-to-blender/references/mcp-backends.md and stay on the backend you picked for the whole task.
What it helps with
- bring four orthographic views onto one scale and center
- repair side and back silhouettes that disagree with the front
- add
Ydepth to a front-locked contour model - align reference planes and cameras before modeling starts
- log a view conflict in the report instead of editing the front
Example prompts for the agent
- Run
scripts/register_views.pyon my view images and build the reference planes at one shared scale - Align the centerline and bbox centers first, then lock the front boundary vertices
- Add
Ydepth from the side and top envelopes and validate all four cameras before export
How to install orthographic-registration
1Skills CLI (Vercel)
npx skills add RobLe3/cc-blender-skill --skill orthographic-registration -a claude-code2Manually: put the files in place
~/.claude/skills/orthographic-registration/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/orthographic-registration/
1Skills CLI (Vercel)
npx skills add RobLe3/cc-blender-skill --skill orthographic-registration -a codex2Manually: put the files in place
.agents/skills/orthographic-registration/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add RobLe3/cc-blender-skill --skill orthographic-registration -a cursor2Manually: put the files in place
.cursor/skills/orthographic-registration/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install RobLe3/cc-blender-skill orthographic-registration2Manually: put the files in place
.github/skills/orthographic-registration/SKILL.md1Skills CLI (Vercel)
npx skills add RobLe3/cc-blender-skill --skill orthographic-registration -a gemini-cli2Platform CLI
gemini skills install https://github.com/RobLe3/cc-blender-skill --path plugin/skills/orthographic-registrationManually: put the files in place
.devin/skills/orthographic-registration/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add RobLe3/cc-blender-skill --skill orthographic-registration -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about orthographic-registration
How do I install orthographic-registration in Claude?
Run npx skills add RobLe3/cc-blender-skill --skill orthographic-registration -a claude-code. The skill lands at ~/.claude/skills/orthographic-registration/SKILL.md.
What does orthographic-registration do?
It registers front, side, back and top orthographic views into one Blender coordinate contract so the silhouettes share a scale and center. It also sets the order for aligning planes and cameras and for adding depth.
What does the skill need to work?
A Blender MCP server. The skill calls its tools for code execution, object summaries, viewport renders and API doc search. Before the first Blender call it asks you to read ../text-to-blender/references/mcp-backends.md and keep the same backend for the task.
Source text
SKILL.md · MIT
--- name: orthographic-registration description: Register front, side, back, and top orthographic reference views into a shared Blender coordinate contract. Use when a 3D model must match multi-view wireframes/templates, when side/back/top silhouettes are off, or before adding depth to a front-locked contour model. allowed-tools: Read Bash Glob Grep mcp__blender__execute_blender_code mcp__blender__get_scene_info mcp__blender__get_objects_summary mcp__blender__get_object_detail_summary mcp__blender__render_viewport_to_path mcp__blender__search_api_docs mcp__blender_official__execute_blender_code mcp__blender_official__get_objects_summary mcp__blender_official__get_object_detail_summary mcp__blender_official__render_viewport_to_path mcp__blender_official__search_api_docs --- ## MCP backend Use when: Multi-view mascot/object reconstruction, depth fitting from side/back/top views, orthographic blueprint alignment, reference planes/cameras with shared scale and centerline. Before the first Blender call, read [../text-to-blender/references/mcp-backends.md](../text-to-blender/references/mcp-backends.md). Prefer official MCP for shared capabilities; preserve the selected backend for the whole task. Tool names in recipes are client/legacy examples: map them through that reference and its agent-host guide. # Orthographic Registration This skill prevents the common failure where the front view looks plausible but the side/back/top views are wrong. ## Coordinate contract - Front view defines `X/Z` silhouette. - Side view defines `Y/Z` depth envelope. - Top view defines `X/Y` spread. - Back view defines rear silhouette/material only; it must not rewrite the front silhouette. ## Workflow 1. Run `scripts/register_views.py` on the available view images. 2. Create orthographic reference planes/image empties with a single shared scale. 3. Align centerline and bbox centers before modeling. 4. Lock front X/Z boundary vertices. 5. Add Y depth from side/top envelopes using modifiers/displacement or vertex groups. 6. Validate all four cameras before export. ## Failure policy If views disagree, front brand read wins. Document the conflict in `registration_report.json`. ## Front-plane rotation rule In this Blender coordinate convention, the front camera looks along the Y axis and the reference silhouette lives in the X/Z plane. Therefore 2D rotations inside the front view are rotations around the **Y axis**, not around Z. Use `rotation_euler = (0, angle, 0)` for 2D component orientation in the front projection. Z rotation spins objects into/out of screen-space incorrectly for X/Z-plane meshes. ## Sources distilled - Blender Image Empty/reference image controls for orthographic blueprints. - OpenCV homography/alignment and geometric transforms for view registration.
© RobLe3 · MIT · RobLe3/cc-blender-skill