AimatraTHIRD-PARTY OPEN SOURCE · MIT
Design a lab holder from measured dimensions
A third-party open-source skill by K-Dense Inc. for turning measured labware dimensions into parametric Python CAD models, fabrication files, and geometry checks.
What can it do?
Provide the dimensions and fabrication process for a simple holder or adapter; the local agent helps create a model, export files, and a report of the checks.
You provide
- A sketch or written brief with dimensions in millimetres, part orientation, mating surfaces, and the intended use.
- The actual labware measurements, printer or machining limits, desired clearance, and any existing trusted CAD source you want to preserve.
You receive
- A parametric Python model with named dimensions, STEP and STL files, and an optional DXF section when a 2D cutting profile is suitable.
- PNG views, geometry and interface-check results, and a provenance manifest recording parameters, versions, and unchecked dimensions.
Documentation reviewed: 2026-10-19. Installation and the complete workflow have not been tested end to end by AimatraLab.
Prepare your project
Open your own project in Codex or Claude Code, with permission to read and write files and run commands. An ordinary web chat may not be able to install a Skill.
- Codex or Claude Code with permission to read and write project files and run local commands. A chat page without file tools cannot install this skill.
- Python 3.11–3.14, build123d 0.13.0, and Matplotlib for PNG snapshots. The upstream setup uses uv and a project virtual environment.
- Measured mating dimensions or the exact vendor drawing, plus the fabrication process, material, and a tolerance budget. Local geometry checks can run offline after setup.
These requests run in your assistant's host environment. This page provides guidance; it does not run AI tasks or collect your project files.
Ask to install
Copy this request into your project conversation. Ask the assistant to review the original Skill and explain any required setup before using it.
Installation request
I am using Codex or Claude Code with filesystem and shell tools. Install Lab Hardware CAD from https://github.com/K-Dense-AI/scientific-agent-skills at commit 593377bb048752abb0152deab9fbb1083a6087df, folder skills/lab-hardware-cad. Read SKILL.md and retrieve the complete skill, including references, scripts, and assets, not just SKILL.md. Preserve the original repository LICENSE.md and K-Dense Inc. attribution. Use the current project's skill directory supported by this host, following existing project configuration. Do not overwrite an existing skill; explain any conflict first. Check required Python, build123d, Matplotlib, and other dependencies and explain proposed environment changes before installing them. Tell me the installation path and how to invoke the skill, then wait for my measurements and design task before running any model.
You can also select the request text above and copy it manually.
Try your first task
Once installation is confirmed, replace the example paths and details with your own files, then send this request in the same project.
First-use request
Use the installed lab-hardware-cad skill to design a simple desktop cuvette holder for a dry lab bench. First ask for my measured outer cuvette dimensions in millimetres, the required number of positions, and my fabrication process and material. Do not treat a 10 mm optical path as the outside width or invent missing dimensions. Explain the clearance and tolerance budget before modeling. Preserve my original sketches and CAD files, and save a new parametric Python model, STEP, STL, PNG snapshots, and a check report in a new output folder. Inspect the geometry and PNG views, state what remains unchecked, and recommend measuring a printed fit coupon before fabricating the full holder.
You can also select the request text above and copy it manually.
Review the result
- A parametric Python model with named dimensions, STEP and STL files, and an optional DXF section when a 2D cutting profile is suitable.
- PNG views, geometry and interface-check results, and a provenance manifest recording parameters, versions, and unchecked dimensions.
Ask your assistant where each output was saved and how it was checked. Compare the result with your original inputs before you use or share it.
Questions & limits
- Source and documentation checked on 2026-10-19; complete runtime workflows have not been tested here.
- A valid CAD solid or passing software check does not prove the fabricated part fits. Measure mating parts and test a fit coupon; process tolerances must be calibrated.
- Material compatibility, optical performance, loads, and instrument clearances need separate real-world checks for the intended use.
- The upstream tools execute Python model files when building them. Read externally supplied model code before running it.
Can I upload files and run this here?
This guide does not provide online execution or file uploads. Work with the files in your own Codex or Claude Code project.
What if my assistant cannot install it?
Confirm that you are in a project conversation with file and command access. Ask the assistant to show the failed step and missing dependency, then use the pinned source below to check the instructions.
Which terms apply to this Skill and its dependencies?
See MIT in the source section below. Dependencies and fonts have their own licences; services and data may have separate terms, access requirements or costs.
Source & version
A third-party open-source selection by AimatraLab. The Skill belongs to its original author; source links go directly to their repository.
- Original author
- K-Dense Inc.
- Repository
- https://github.com/K-Dense-AI/scientific-agent-skills
- Selected version
- 1.5
- Fixed commit
593377bb048752abb0152deab9fbb1083a6087df- License
- MIT
- Documentation checked
- 2026-10-19