# AI use cases for Multi Agent CAD (Pan-Chera/Multi-Agent-CAD)

Repo: [Pan-Chera/Multi-Agent-CAD](https://github.com/Pan-Chera/Multi-Agent-CAD) | Use cases: 5 | Page: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/

MAC is a Python framework that turns natural-language design briefs into editable CAD parts and multi-part assemblies, exported as STEP, STL, and GLB files.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Generate an editable printable part from a text brief: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/generate-printable-part-from-brief/
- Run a local browser UI to generate and download parts: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/draft-browser-ui-part-run-for-team/
- Review generated CAD with QA reports and repair history: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/audit-generated-cad-with-qa-history/
- Prototype an articulated assembly and export URDF: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/prototype-assembly-with-urdf-export/
- Have a coding agent generate part variants from briefs: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/agent-generates-part-variants-from-briefs/

## How to generate an editable printable part from a text brief with Multi Agent CAD

Repo: [Pan-Chera/Multi-Agent-CAD](https://github.com/Pan-Chera/Multi-Agent-CAD) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README

Turn a plain-language part description into STEP, STL, and GLB files you can open in CAD or print.

**Who it is for:** A maker or product designer who needs a first CAD model of a simple printable part and does not want to write CAD code.

**The problem:** Writing parametric CAD from scratch takes time, and a text description alone does not produce a file you can edit or print.

**The outcome:** An editable part exported as STEP, STL, and GLB, with the generated Python and QA results available for review.

### Steps

1. `git clone https://github.com/Pan-Chera/Multi-Agent-CAD.git`
2. `cd Multi-Agent-CAD`
3. `conda env create -f environment.yml`
4. `conda activate multi_agent_cad`
5. `pip install --no-deps "aider-chat==0.82.3"`
6. `export DASHSCOPE_API_KEY="your-key"`
7. Set `USER_REQUEST` in `multi_agent_cad/config.py` to your part brief, then run `python -m multi_agent_cad.graph`

### Prompt to paste into your agent

```text
Use the Multi-Agent-CAD repo (github.com/Pan-Chera/Multi-Agent-CAD). Set USER_REQUEST in multi_agent_cad/config.py to the brief <part description>, then run python -m multi_agent_cad.graph. Report the exported STEP, STL, and GLB paths and confirm the QA result before printing.
```

### Caveat

The README says generated geometry should be validated before manufacturing use, so check dimensions and fit before printing.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/generate-printable-part-from-brief/

## How to run a local browser UI to generate and download parts with Multi Agent CAD

Repo: [Pan-Chera/Multi-Agent-CAD](https://github.com/Pan-Chera/Multi-Agent-CAD) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README

Start the MAC web UI on your own machine to generate parts with a 3D preview and download results.

**Who it is for:** A designer or small team member who prefers a visual interface to editing config files and running commands.

**The problem:** Running a CLI pipeline and finding outputs is awkward for people who want to type a brief, watch progress, and preview the model.

**The outcome:** A local web page where you submit a brief, see a 3D preview, and download the generated part files.

### Steps

1. `git clone https://github.com/Pan-Chera/Multi-Agent-CAD.git`
2. `cd Multi-Agent-CAD`
3. `conda env create -f environment.yml`
4. `conda activate multi_agent_cad`
5. `pip install -e ".[web]"`
6. `python -m multi_agent_cad.web`
7. Open the local address the server prints in your browser, enter your brief and key, and run a job (inferred, check the README)

### Prompt to paste into your agent

```text
Use the Multi-Agent-CAD repo (github.com/Pan-Chera/Multi-Agent-CAD). Start the web UI with python -m multi_agent_cad.web on 127.0.0.1, submit the brief <part description>, and confirm a downloadable result with a 3D preview appears.
```

### Caveat

The README warns the UI executes generated Python with your user account's permissions, so keep it on loopback unless you understand the network risk.

Risk, credentials: Enter your model API key only into the local UI form, and never expose the server beyond your own machine, because the UI runs generated Python with your user account's permissions.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/draft-browser-ui-part-run-for-team/

## How to review generated CAD with QA reports and repair history with Multi Agent CAD

Repo: [Pan-Chera/Multi-Agent-CAD](https://github.com/Pan-Chera/Multi-Agent-CAD) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README

Inspect the brief, plan, generated Python, measurements, and repair steps behind each generated part.

**Who it is for:** A mechanical engineer who must sign off on AI-generated geometry and needs to see how it was made before trusting it.

**The problem:** A finished model alone does not show whether dimensions were checked or how failures were fixed, which makes review slow.

**The outcome:** A clear record of the structured brief, geometry plan, generated code, measurements, QA report, and repair history for one part.

### Steps

1. `git clone https://github.com/Pan-Chera/Multi-Agent-CAD.git`
2. `cd Multi-Agent-CAD`
3. `conda env create -f environment.yml`
4. `conda activate multi_agent_cad`
5. `pip install --no-deps "aider-chat==0.82.3"`
6. `export DASHSCOPE_API_KEY="your-key"`
7. Run `python -m multi_agent_cad.graph` with your brief, then open the run's QA report and repair history (inferred: check the output location set in `multi_agent_cad/config.py`)

### Prompt to paste into your agent

```text
Use the Multi-Agent-CAD repo (github.com/Pan-Chera/Multi-Agent-CAD). Run the single-part workflow for the brief <part description>, then summarize the QA report and repair history and list any measurements that do not match the brief.
```

### Caveat

The README describes audit artifacts but the exact output folder layout is in the single-part workflow documentation, so confirm where files land before relying on the path.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/audit-generated-cad-with-qa-history/

## How to prototype an articulated assembly and export URDF with Multi Agent CAD

Repo: [Pan-Chera/Multi-Agent-CAD](https://github.com/Pan-Chera/Multi-Agent-CAD) | Difficulty: Advanced | Time: 2 hours | Basis: From the README

Describe a multi-part mechanism and get a generated assembly with a URDF file for simulation work.

**Who it is for:** A robotics or mechanism developer who wants a quick articulated concept to test in a physics simulator.

**The problem:** Modeling each linked component, its joints, and a simulator description by hand is slow for early concept checks.

**The outcome:** An experimental multi-part assembly with joint interfaces and a URDF export for downstream robotics or physics simulation.

### Steps

1. `git clone https://github.com/Pan-Chera/Multi-Agent-CAD.git`
2. `cd Multi-Agent-CAD`
3. `conda env create -f environment.yml`
4. `conda activate multi_agent_cad`
5. `pip install --no-deps "aider-chat==0.82.3"`
6. `export DASHSCOPE_API_KEY="your-key"`
7. Read [assembly workflow documentation](mac_assembly/README.md) and follow its steps to describe the mechanism, generate the assembly, and export URDF

### Prompt to paste into your agent

```text
Use the Multi-Agent-CAD repo (github.com/Pan-Chera/Multi-Agent-CAD). Follow mac_assembly/README.md to generate an assembly from the mechanism description <mechanism description> and export URDF. Report the URDF path and list the joints, then check that each joint limit matches the brief.
```

### Caveat

The assembly workflow is an experimental technology preview, and the README says to validate generated geometry and joints before manufacturing or simulation use.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/prototype-assembly-with-urdf-export/

## How to have a coding agent generate part variants from briefs with Multi Agent CAD

Repo: [Pan-Chera/Multi-Agent-CAD](https://github.com/Pan-Chera/Multi-Agent-CAD) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README

Let an agent set the brief, run the pipeline, and return exported part files for each variant.

**Who it is for:** An engineer who uses a coding agent for routine work and wants it to produce several CAD variants from a list of briefs.

**The problem:** Preparing and running each variant by hand is repetitive when only dimensions or small features change.

**The outcome:** A set of generated part files, one per brief, with the agent reporting each output path and QA result.

### Steps

1. `git clone https://github.com/Pan-Chera/Multi-Agent-CAD.git`
2. `cd Multi-Agent-CAD`
3. `conda env create -f environment.yml`
4. `conda activate multi_agent_cad`
5. `export DASHSCOPE_API_KEY="your-key"`
6. For each brief, set `USER_REQUEST` in `multi_agent_cad/config.py` and run `python -m multi_agent_cad.graph`, then collect the exported files (inferred, check the README)

### Prompt to paste into your agent

```text
Use the Multi-Agent-CAD repo (github.com/Pan-Chera/Multi-Agent-CAD). For each brief in <list of briefs>, set USER_REQUEST in multi_agent_cad/config.py, run python -m multi_agent_cad.graph, and return the STEP and STL paths with QA results. Verify that each output matches its brief dimensions before reporting.
```

### Caveat

Each run uses model calls billed to your provider, and the README's cost figures come from its own benchmark, not from your runs.

Risk, credentials: The agent needs your model API key in its environment, so limit which agent runs have access to it.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/pan-chera-multi-agent-cad/agent-generates-part-variants-from-briefs/