Solidworks Automation logo

Solidworks Automation

Community
delancy827
solidworks-automation

用户需要用 Python/C#/VBA 连接 SolidWorks API 自动建模、装配、工程图、仿真等时调用本技能。

Overview

Publisherdelancy827
Repositorysolidworks-skills
Skill namesolidworks-automation
Stars
150
Forks
15
Bundled files
13
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 13 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by delancy827 on GitHub. Read the source before you install it.

Installation

Install the Solidworks Automation AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/delancy827/solidworks-skills/tree/main/solidworks-automation
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/delancy827/solidworks-skills.git /tmp/solidworks-skills
mkdir -p .claude/skills
cp -r /tmp/solidworks-skills/solidworks-automation .claude/skills/solidworks-automation
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Solidworks Automation in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Solidworks Automation on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Solidworks Automation is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

SolidWorks Automation

Use this skill when the task requires executable SolidWorks automation code or a live COM session for parts, assemblies, drawings, Simulation, Flow Simulation, sheet metal, weldments, molds, or tooling. For design-only decisions or parameterization methodology without implementation, use sw-designer instead.

Operating rules

  • Reuse an active SolidWorks instance with GetActiveObject("SldWorks.Application"); only fall back to Dispatch when no usable instance exists.
  • Reuse sw.ActiveDoc when it matches the task. Create a new document only when explicitly required or no suitable document exists.
  • Wrap temporary documents in try...finally and close them by title; do not leave orphan windows or COM references.
  • Treat units as meters inside the API. Keep user-facing dimensions in millimeters and convert explicitly.
  • Before any mutation, capture a geometry fingerprint: feature count, body count, bbox, volume, and relevant dimensions.
  • After each mutation, force a rebuild, re-read the fingerprint, and assert the expected change. Do not report success from a return value alone.
  • If an assertion fails, stop the current strategy, print the traceback and current feature-tree state, and propose a concrete fallback. Do not silently catch the failure.
  • Do not guess API names, argument counts, or COM null handling. Probe the live object with dir(), get_com_member, or a tiny subprocess before using an uncertain API.
  • Keep real project files and generated models under F:\codexspace\...; do not copy full projects to the Desktop or C drive.

Recommended workflow

  1. Preflight capability. State whether the current runtime can inspect images/PDFs and whether SolidWorks is available. For drawing-driven work, identify the drawing source and who or what performed visual verification.
  2. Build a dimension ledger. Separate exact, inferred, designer-choice, and unsupported dimensions. Bounding-box matching is not sufficient; account for recesses, lands, grooves, PCD holes, chamfers, and secondary features.
  3. Choose an execution channel.
    • Prefer Python COM for connection, sketching, additive extrudes, measurements, and file I/O.
    • Prefer one sketch with outer and inner closed loops plus FeatureExtrusion2 for through holes and rings.
    • Use C# or a generated VBA macro for methods that Python COM cannot call reliably, especially complex cut features with more than 12 parameters.
  4. Implement in the project workspace. Keep builder code modular, parameter-driven, and testable without loading the full manual.
  5. Execute and validate. Compile the script first. If SolidWorks is available, run it, force a rebuild, then assert feature/body counts, bbox, volume, and output-file existence. Use screenshots or rendered views when geometry intent matters.
  6. Report truthfully. Distinguish completed geometry, markers-only holes, approximations, and unsupported features. Never describe an uncut marker as a through hole.

Known SW 2024 COM shortcuts

  • SelectByID2 has 9 arguments; the Callout argument must be a real COM null variant, typically VARIANT(VT_DISPATCH, None).
  • FeatureExtrusion2 uses 23 arguments on this install.
  • EditRebuild3 may be a property instead of a method. ForceRebuild3(False) is the safer rebuild call.
  • FeatureCut, FeatureCut3, and FeatureCut4 are fragile from Python COM. Use additive geometry, multi-contour sketches, C#, or VBA rather than pretending a None return means success.
  • LoadFile4 may return (doc, status); unwrap it before using the document.
  • Extension.SaveAs needs COM variants for export data and by-ref error/warning values.
  • For assemblies, preload components, insert at process coordinates, then mate using stable named entities or SelectByRay; verify BOM count and envelope before attempting perfect mates.

References

Read these only when the current task needs them.

  • references/automation-manual.md — the full 46-section manual. Search it first with rg -n "keyword" references/automation-manual.md, then read the narrow line range. Avoid loading the whole file by default.
  • references/drawing-to-3d-lessons.md — DXF/PDF-to-part rules, secondary-feature checks, volume calibration, and ground-truth STEP comparison.
  • references/materials-forming-mold-lessons.md — stamping edge sizing, die-cast shrink direction, process metadata, moldbase checks, and mold assembly QA.
  • references/assembly-debugging-lessons.md — repeated-instance counting, transformed hole centers, and render-based assembly validation.
  • references/com-dualinterface-quirks.md — COM 双接口怪癖实证手册(2026-10 单机全流程实战沉淀)。当动态 COM 调用报"找不到成员"、拉伸静默返回 None、AddComponent5 静默失败、ComposeTransform 摆位不对、截图视图切不过去时,先读它。配套可运行流水线见 examples/verified/。
  • examples/01_basic_part.py
  • examples/02_assembly.py
  • examples/03_drawing.py
  • examples/04_simulation.py

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Solidworks Automation AI skill do?

用户需要用 Python/C#/VBA 连接 SolidWorks API 自动建模、装配、工程图、仿真等时调用本技能。

Why use Solidworks Automation on TypingMind?

Because you install it once and use it with any model. Solidworks Automation is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Solidworks Automation in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/delancy827/solidworks-skills/tree/main/solidworks-automation. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Solidworks Automation?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Solidworks Automation?

As many as you like. As long as a model supports skills, you can use Solidworks Automation with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Solidworks Automation AI skill free?

Yes. It is published on GitHub by delancy827 under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇