
jmwright/cadquery-llm-skill
Productivity workflowこのページは AI により自動翻訳されています。技術的な詳細は原文をご確認ください。 原文を見る
A skill to help LLMs understand CadQuery and BRep modelling better
概要
A skill that helps LLMs write correct, idiomatic CadQuery code. LLMs regularly get CadQuery wrong - defaulting to Boolean/CSG thinking instead of CadQuery's BRep-native approach, misusing selectors, building in global coordinates, and reaching for .union() / .cut() when a workplane operation would be cleaner. This skill gives Claude Code the context it needs to do better. - - dense quick-reference: BRep mindset, both APIs, selector cheat sheet, critical anti-patterns - - deep-dive files on BRep vs CSG, workplanes, selectors, and the Free Function API - - verified common patterns and anti-patterns with explanations - - annotated runnable scripts - - local snapshot of the CadQuery RST documentation for offline grepping - - Ollama model definition with the skill pre-loaded as a system prompt - - utility scripts, including one to regenerate a full-content Modelfile from the markdown sources Clone the repo somewhere accessible on your machine: Prefix any CadQuery prompt with @ to load...
README
以下は原文 README の抜粋です。原文を見る。
cadquery-llm-skill
A skill that helps LLMs write correct, idiomatic CadQuery code.
LLMs regularly get CadQuery wrong - defaulting to Boolean/CSG thinking instead of
CadQuery’s BRep-native approach, misusing selectors, building in global coordinates,
and reaching for .union() / .cut() when a workplane operation would be cleaner.
This skill gives Claude Code the context it needs to do better.
What’s Included
SKILL.md- dense quick-reference: BRep mindset, both APIs, selector cheat sheet, critical anti-patternsconcepts/- deep-dive files on BRep vs CSG, workplanes, selectors, and the Free Function APIpatterns/- verified common patterns and anti-patterns with explanationsexamples/- annotated runnable scriptsdocs/- local snapshot of the CadQuery RST documentation for offline greppingModelfile- Ollama model definition with the skill pre-loaded as a system promptscripts/- utility scripts, including one to regenerate a full-content Modelfile from the markdown sources
Installation
Clone the repo somewhere accessible on your machine:
git clone https://github.com/jmwright/cadquery-llm-skill.git ~/cadquery-llm-skill
Usage
Option 1 - Reference in a prompt (one-off)
Prefix any CadQuery prompt with @ to load the skill file:
@~/cadquery-llm-skill/SKILL.md
Write a CadQuery script that creates a flanged pipe fitting.
Claude Code will read SKILL.md before responding. For deeper questions, reference
specific concept files:
@~/cadquery-llm-skill/SKILL.md
@~/cadquery-llm-skill/concepts/selectors.md
Why is my selector not picking the right face?
Option 2 - Add to a project’s CLAUDE.md (persistent)
Add the following to your project’s CLAUDE.md file so Claude Code loads the skill
automatically for every CadQuery task in that project:
## CadQuery Skill
This project uses CadQuery. Before writing any CadQuery code, read:
@~/cadquery-llm-skill/SKILL.md
For deeper reference:
- Selectors: @~/cadquery-llm-skill/concepts/selectors.md
- Workplanes: @~/cadquery-llm-skill/concepts/workplanes.md
- Free Function API: @~/cadquery-llm-skill/concepts/free-function-api.md
- Anti-patterns: @~/cadquery-llm-skill/patterns/anti-patterns.md
When unsure about a method, grep the local docs:
grep -rn "method_name" ~/cadquery-llm-skill/docs/
Adjust the path to match where you cloned the repo.
Option 3 - Global CLAUDE.md
To enable the skill for all projects, add the same block to ~/.claude/CLAUDE.md.
Option 4 - Ollama / Open Models
The repo includes a Modelfile that can be used to create a local cadquery-coder
model in Ollama with a focused version of the skill pre-loaded as a system prompt.
ollama create cadquery-coder -f Modelfile
ollama run cadquery-coder
Model requirements: This skill was designed and tested primarily with Claude Code.
推奨ツール
別のキーワードを試すか、フィルタを外してください。
インストール
npx skillfish add jmwright/cadquery-llm-skill