prompt-driven-design

A guide for describing SolidWorks designs in plain language and having an AI translate them into SolidWorks tool calls. SolidWorks is software for computer-aided mechanical design.

In plain words
What is it for?
Use it to describe dimensions and features, review the proposed tool-call sequence, execute it, and check the resulting part file.
Why use it?
It helps you create or inspect SolidWorks parts through a described plan instead of manually performing every menu operation.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/andrewbartels1/solidworksmcp-python/prompt-driven-design
Clone the repo
git clone --depth 1 https://github.com/andrewbartels1/SolidworksMCP-python
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,406 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.03406
Opus 5 $0.00000 $0.01703
Sonnet 5 $0.00000 $0.00681
Haiku 4.5 $0.00000 $0.00341

Measured 2d ago against content hash 67af58e81a1a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

prompt-driven-design scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

docs/agents/prompt-driven-design.md · 378 lines

How it starts

The opening of the file, as written. The whole thing — 378 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Prompt-Driven Design with SolidWorks MCP

You know SolidWorks. This guide shows you how to describe what you want to build in plain language, let an LLM translate that into the right tool calls, and have SolidWorks build it — without clicking through menus.

The examples below use real SolidWorks sample parts as targets. Because the finished .SLDPRT files already exist, you can open them and check the feature tree to confirm whether your prompts produced the right result. That makes them ideal for learning the workflow before applying it to your own designs.


What You Need

  • SolidWorks installed and open
  • MCP server running (.\dev-commands.ps1 dev-install then .\dev-commands.ps1 dev-run if needed)
  • Claude Code (this tool) or VS Code with Copilot Chat
  • Python environment activated: .\.venv\Scripts\python.exe

The Core Workflow

Every design session follows the same loop:

flowchart LR
    D(["1 - Describe<br/>What to build<br/>+ dimensions"])
    P(["2 - Plan<br/>LLM proposes<br/>MCP call sequence"])
    R(["3 - Review<br/>You approve<br/>the plan"])
    E(["4 - Execute<br/>Claude Code runs<br/>MCP tools"])
    C(["5 - Check<br/>Visual or<br/>feature-tree"])
    I(["6 - Iterate<br/>Refine dimensions,<br/>add features"])

    D --> P --> R --> E --> C --> I
    I -->|new idea or fix| D

    style R fill:#f0f4ff,stroke:#4a6cf7
    style C fill:#f0f4ff,stroke:#4a6cf7

Steps 3 (Review) and 5 (Check) are yours. The LLM handles 2 and 4.


Step 1 — Describe What You're Building

Be specific about shape and size. Adjectives like "wide" or "thin" are ambiguous. Geometry and dimensions are not.

Good description

I want to create a paper airplane shape.
It should be a flat delta-wing profile — like looking down at a paper airplane from above.
The nose points right at (80, 0), the two wing tips are at (0, 60) and (0, -60),
and everything connects back to the origin.
Extrude it 0.5 mm — just enough thickness to represent folded paper.

Read the full file on GitHub · 378 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 378 lines · 0 tokens per session scan A 67af58e81a1a

Subscribe to this mod's changes

prompt-driven-design is an agent published in the GitHub repository andrewbartels1/SolidworksMCP-python (65 stars, last pushed 6d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,406 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories