Full Version: mcp-bridge-for-moi: let an AI agent model in MoI while you watch

Show messages:  1-18  19-37

Hi Burr,

> Why does “Claude” need MoI?

I was fooling with Grok some months back, asking it to produce some dimensioned, scaled architectural models. It quickly turned into a mess, so I began asking "it" about its underlying tools and abilities. What I learned was that none of the then current AIs had integral geometry kernels though there were some efforts underway to communicate with some CAD packages (typically work being done by the CAD vendors).

At that point I shrugged and went back to using MoI as I normally do because for me, drawings without scale and precision are not useful.

I haven't visited Gabriel's github yet, but am a bit curious about how it controls MoI.

--Larry

---

Hey Larry,
what you're saying is absolutely right, and it's exactly why the MCP approach works differently.

Most commercially available AI models still have no geometry kernel, and mcp-bridge-for-moi doesn't pretend otherwise. It doesn't generate geometry at all. All the geometry is made by MoI's own kernel, in your open document, through MoI's scripting API. The AI's only job is to decide which MoI command to run and with what numbers — the same decisions you make when you type a value into a command's input field.

How it works in practice

Two pieces: a small script that sits in MoI's startup folder and connects out over a local WebSocket, and an MCP server that your AI client (Claude Code, Claude Desktop, anything that speaks MCP) launches. The server talks to the agent on one side and to MoI on the other. It listens on 127.0.0.1 only, with a token regenerated each start.

The agent gets a handful of tools. The important one runs JavaScript inside MoI, which is how it calls MoI's geometry factories — the same 110 commands the UI drives (box, loft, revolve, fillet, boolean, …). 103 of them can be driven from a script; the other 7 need an interactive viewport pick or produce nothing headless, and are documented as such. Every one of the 103 is committed with real inputs and its output checked by an automated test run.

The rest of the tools are: list the scene (id, type, whether an object is a closed solid, bounding size, document units), read your selection, set the selection or the view, render a viewport image so it can look at what it just made, and export to STEP/OBJ/STL etc.

On scale and precision specifically

- Every factory input is a number in document units. A box 3600 × 240 × 2700 mm is that, exactly, because MoI built it.
- If the open document has no unit system, modelling and export are refused with a no_units error until you pick one.
- Exports go through MoI's own translators, so a STEP out of it is a MoI STEP.

So the failure mode you hit with Grok — plausible-looking nonsense with no kernel underneath — isn't really available here. What can go wrong is different and more mundane: the agent picks the wrong command, mis-orders an operation, or misjudges a dimension it inferred rather than was given. You see it happen live in your viewport.

Cheers,
Gabriel

---

@ Frenchy Pilou
Sorry, Pilou. I misunderstood you in the first place — I thought you were talking about controlling an AI via voice commands.
I guess your solution breakdown (especially using macro tools) would work.

Cheers,
Gabriel

---

@Matadam

Matadam, I'm revising what I said before: although the MCP server can't build any scripts on its own, it can be used for creating and evaluating scripts.
Therefore, an AI agent building JavaScript would evaluate its functions through the MCP server.
I did a quick test yesterday and it worked.

Hope it helps!
Cheers, Gabriel

---

@Gabriel

No Problem ! :)

Voice is very practical for people who can't type on a keyboard!

---

Thanks for the additional explanations Gabriel. At this point I've not done any experimenting with agentic-AI, but who knows, maybe your project will nudge me to do so. In the meantime cheers for the unique idea and experiment which you have made available for others to learn with and from.

--Larry

---

@Larry

Thank you so much for your kind words, Larry.
A small quick tip for you and anyone else who wants to try AI-assisted MoI production: I have successfully tried the MCP with Google's 'Antigravity'.
'Antigravity' is free for anyone with a Gmail account to try. Its capabilities are limited. Nevertheless, it gives you something to experiment with.

Cheers,
Gabriel

---

@Gabriel,
I checked your mcp Server on both actual macOS and moi V5. Works Like charming hell. Thank you very much. The only thing missing in v5 are the 5 new factories .

---

Hey binfordboy,

Thank you for trying it out :)
Many thanks for testing it on MacOS! I'll update the README accordingly

I'll check out V5 when I have a spare moment. I'll have to implement a fallback mechanism for V4, though.

Cheers,
Gabriel

---

Excellent! Great Job, this is going to get better and better!

---

So as a complete newbie here:

I download/copy

Bridge.js
node-v22.22.2-x64.msi

Put the Bridge file into AppData/MOI

install the msi file

open MOI 4.0

Is that really it?

---

@ ado1

Hey ado1,
if the MCP server is connected properly to your AI agent (Claude etc.), you can do a quick check by opening a fresh scene in MoI and writing in your AI session:
call get_scene

IF the agent responds with something like "Scene is emtpy", then you're all set. Otherwise something is missing. You can also check the MCP installation by typing /mcp in your session - it must show you the list of installed MCP servers, among them the moi server.

I'll try to find a minute to make a short video on how to install the server.

Cheers,
Gabriel

---

OK thx.

I've only got Win7 Pro on my main setup so the node won't install but I've got a Tablet I can give it a shot on.
That's the first time I've ever had cause to regret not being more up-to--date

Which reminds me if Mr Gibson is around.

What's the situation with my Moi 4 License Mr Gibson, Can I use it on a second computer for testing this AI out?

---

You can have 2 or more computers with Moi installed on them BUT...
...you must use only ONE computer in the same time for Moi! ;)

---

Awesome.
Thanks for that.

---

Ado1 direct your ai to the gitrepository and Tell her to install and Check everything on your Computer is the easy way

---

Hi everyone,
an important notice:
mcp v. 0.2.0 has been released

The full change log can be found here: Changelog v 0.2.0

what changed? => the export function is now available in the MCP.

Please update.
A short description on that can be now found on the repo main page.

Cheers,
Gabriel

---

Hey everyone,
v. 0.4.0 is out
This release only changes the server. The MoI side is unchanged, so you don't need to reinstall the bridge. Pull, rebuild, and restart your AI client.

What's new:

• Better help on MoI objects. moi_factory_help can now describe objects, edges and styles, which MoI's own API doesn't document. It works without a running MoI session.
• Clearer hints when a script goes wrong.
• If a script returns nothing because it has no return, the reply now says so.
• If a script fails on a large model, the error stays short: it lists up to 10 object ids and gives a count for the rest.
• Nine more MoI scripting pitfalls are explained to the agent. These include commit() return values, mesh import, join returning one object per connected group, and isClosed on breps. The whole list is now easier for the agent to read.
• Units. On a document with no unit system, the agent may set millimeters on its own when it's only testing. Otherwise it still asks you first.

Full changelog and download:
Changelog mcp-bridge-for-moi 0.4.0

Cheers,
Gabriel

---

Hey everyone,

v. 0.5.0 is out: see what the tools cost your agent

This release is about token cost. It changes nothing in the MoI bridge, so updating the server is enough.

• New tokens command. "mcp-bridge-for-moi tokens" estimates how many tokens each tool definition costs an agent, and the total. It needs no MoI session. Add --json to print the definitions as the server sends them.
• Token use section in the README. It lists the cost tool by tool and explains how to check it in your own client. With every definition loaded up front, the estimate is about 3,900–4,400 tokens. Claude Code's /context counted about 5,700 in my test. Clients that load definitions on demand pay less.
• A size guard. A test fails ifst a set size, so the cost can't creep upunnoticed.
• moi_eval description trimmed.de's 2,048-character cap on tooldescriptions, so its ending was cut off. The MoI scripting traps and helper details now live inmoi_factory_help('traps') and moi_factory_
• Pointer on failure. When a moi_eval fails with moi_error, the reply now points to moi_factory_help('traps').

Release notes and download: v0.5.0 on GitHub'

Cheers,
Gabriel

---

Show messages:  1-18  19-37