"Use for IKPy transformation-matrix checks, geometry helpers,
Scanned 9/8/2026
Install to Claude Code
npx -y skills add VectorSpaceLab/AREX-Skill --skill visualization-geometry --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Visualization Geometry?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/vectorspacelab-visualization-geometry)More formats (shields.io, HTML) on the badges page.
---
name: visualization-geometry
description: "Use for IKPy transformation-matrix checks, geometry helpers,
headless 3D chain plots, target and frame overlays, and URDF tree
visualization; do not use it to build chains, solve FK/IK, or control
hardware."
disable-model-invocation: true
metadata:
disco-role: operating
license: Apache 2.0
---
# Visualization and geometry
Use this sub-skill when a Researcher needs to inspect IKPy transforms or create a
visual diagnostic of a chain without connecting to a robot. Start with the
matrix and shape rules in [the API reference](references/api-reference.md),
then select a procedure from [the workflows](references/workflows.md). Keep
long-running or repeatable plots headless and save them to an explicit local
output path.
## Route by responsibility
- Route chain construction, link selection, forward kinematics, inverse
kinematics, and joint-vector semantics to `chain-kinematics`.
- Route URDF/MJCF model loading and parser behavior to `robot-model-import`.
This sub-skill only explains the `get_urdf_tree` visualization utility after
a model path and root link are already known.
- Use the bundled `scripts/smoke_plot.py` for a dependency and save-path smoke
check. It uses an inline toy chain and never reads a robot file or talks to
hardware.
- Treat rendered images and DOT/PDF output as diagnostic evidence, not as a
motion command, simulator state, or proof that a physical robot is safe.
## Operating sequence
1. Establish whether the requested result is a numeric transform inspection, a
Matplotlib image, or a Graphviz tree; they have different optional
dependencies and failure modes.
2. Check matrix dimensions and the RPY multiplication order before rendering.
Inspect the numeric frame first, then use a plot to make frame placement or
target error easier to see.
3. In headless environments, select the Matplotlib `Agg` backend before
importing `ikpy.utils.plot`; call `Chain.plot(..., show=False)` and save the
retained figure explicitly.
4. For a URDF tree, provide an existing URDF path and an exact root link name;
inspect the returned DOT/tree objects before asking Graphviz to render.
5. Record output paths, dependency availability, numerical checks, and any
clipping or convention assumptions in the downstream experiment log.
See:
- [references/api-reference.md](references/api-reference.md) for public
functions, return shapes, conventions, and optional dependencies.
- [references/workflows.md](references/workflows.md) for transform checks,
headless plots, target/intermediate-frame overlays, and DOT rendering.
- [references/troubleshooting.md](references/troubleshooting.md) for import,
backend, shape, plotting, and safety failures.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!