# AI use cases for Ldraw Nova (anteloc/ldraw-nova) Repo: [anteloc/ldraw-nova](https://github.com/anteloc/ldraw-nova) | Use cases: 4 | Page: https://dedupedai.com/use-cases/anteloc-ldraw-nova/ ldraw-nova is a Python toolkit that lets an AI agent build LEGO models from JSON plans and LDraw parts, then generates editable MPD files, checks, previews and parts lists. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Review a sample LEGO model with previews and BOMs: https://dedupedai.com/use-cases/anteloc-ldraw-nova/review-sample-lego-model-previews/ - Build an editable LEGO MPD from a JSON plan: https://dedupedai.com/use-cases/anteloc-ldraw-nova/build-lego-model-from-plan-with-agent/ - Export an LDraw model to GLB for Blender: https://dedupedai.com/use-cases/anteloc-ldraw-nova/export-lego-model-glb-for-blender/ - Run the Ldraw Nova web app with Docker: https://dedupedai.com/use-cases/anteloc-ldraw-nova/run-ldraw-nova-web-app-locally/ ## How to review a sample LEGO model with previews and BOMs with Ldraw Nova Repo: [anteloc/ldraw-nova](https://github.com/anteloc/ldraw-nova) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Run the bridge example through review and get PNG previews, a validation report and parts lists. **Who it is for:** A LEGO hobbyist or developer who wants to see what the Ldraw Nova review pipeline produces before using it on their own designs. **The problem:** Setting up an LDraw-based model pipeline is hard to judge without seeing a working example. **The outcome:** A review folder with previews, the model, a validation report, a Python BOM, a LeoCAD BOM comparison and a manifest. ### Steps 1. Install Python 3.12 or newer, Poppler and LeoCAD, and get an LDraw parts library on disk. 2. Set the library path with `export LDRAW_DIR=/path/to/ldraw`. 3. Run `./setup.sh` from the repository root. 4. Run `./ldraw-agent doctor` to confirm local dependencies are found. 5. Run `./ldraw-agent review examples/bridge.plan.json --outdir output/bridge-review`. 6. Open the PNGs in output/bridge-review/previews/ and read the validation report in the same folder. ### Prompt to paste into your agent ```text In the ldraw-nova repository, run ./ldraw-agent doctor, then ./ldraw-agent review examples/bridge.plan.json --outdir output/bridge-review. Report the artifacts written to output/bridge-review and summarise any warnings or skipped checks in the validation report. Verify that the Python BOM and the LeoCAD BOM comparison both exist before reporting success. ``` ### Caveat Passing checks are evidence, not proof that a model is physically buildable, and the README says visual inspection stays required. 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/anteloc-ldraw-nova/review-sample-lego-model-previews/ ## How to build an editable LEGO MPD from a JSON plan with Ldraw Nova Repo: [anteloc/ldraw-nova](https://github.com/anteloc/ldraw-nova) | Difficulty: Advanced | Time: 2 hours | Basis: From the README Have an agent turn a model idea into a JSON plan, then build and review an editable MPD. **Who it is for:** A LEGO builder or designer who has a model idea and wants an AI agent to draft the construction as an editable file. **The problem:** Turning a rough model idea into a checked, editable LDraw file takes many manual placement and check steps. **The outcome:** An editable MPD model, a reviewed set of previews and checks, and the source plan kept for later repair. ### Steps 1. Give an agent your model idea together with instructions.md from the repository. 2. Have the agent write a JSON plan following docs/agent/plans.md, using Python where calculations or loops help. 3. Run `./ldraw-agent build output/my-model.plan.json` to write output/my-model.mpd once source and geometry checks pass. 4. Run `./ldraw-agent review output/my-model.mpd` and open the previews. 5. Repair the source plan where checks or previews show problems, then repeat the build and review. ### Prompt to paste into your agent ```text Using the ldraw-nova repository and its instructions.md, write a JSON plan for and save it as output/.plan.json. Run ./ldraw-agent build on that plan, then ./ldraw-agent review on the generated MPD. Report the path of the editable MPD, the checks that passed or were skipped, and the preview folder; verify that the MPD exists before reporting. ``` ### Caveat The checks do not certify collision freedom, strength, motion or real part and colour availability, so every design needs human review. 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/anteloc-ldraw-nova/build-lego-model-from-plan-with-agent/ ## How to export an LDraw model to GLB for Blender with Ldraw Nova Repo: [anteloc/ldraw-nova](https://github.com/anteloc/ldraw-nova) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Convert an MPD model to a GLB file with the local CLI so it can be opened in Blender. **Who it is for:** A 3D artist or LEGO fan who wants to bring an Ldraw Nova model into Blender for lighting, camera work or rendering. **The problem:** LDraw MPD files are not directly usable in Blender, and manual conversion is slow. **The outcome:** A GLB file of the model, while the editable MPD stays the authoritative source. ### Steps 1. Complete the local setup with `./setup.sh` and confirm the converter is installed as mpd2glb.sh. 2. Choose an existing MPD file, such as one written by the build step. 3. Run `./prepare-glb.sh --file MODEL.mpd MODEL.glb` with your file names. 4. Import the resulting GLB file into Blender. 5. Make any model edits in the MPD and export again, rather than editing the GLB. ### Prompt to paste into your agent ```text In the ldraw-nova repository, run ./prepare-glb.sh --file .mpd .glb to create a GLB for Blender. Report the GLB path and file size. Verify that the GLB file exists and that the source MPD was not changed. ``` ### Caveat The README describes this export as using the installed mpd2glb.sh converter, so it will fail until that converter is installed and on the expected 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/anteloc-ldraw-nova/export-lego-model-glb-for-blender/ ## How to run the Ldraw Nova web app with Docker with Ldraw Nova Repo: [anteloc/ldraw-nova](https://github.com/anteloc/ldraw-nova) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Start the browser app with 3D and VR views on localhost using Docker Compose. **Who it is for:** A builder who wants interactive 3D and VR views of models, including a Meta Quest 3 headset, on a trusted home network. **The problem:** Viewing generated models in 3D and VR needs a running app and a matching set of repositories. **The outcome:** The web app running at localhost:8765, with HTTPS on port 8443 for VR, and conversation history available. ### Steps 1. Install Git and Docker. 2. Clone matching release tags side by side with `git clone --branch v0.6.0 https://github.com/anteloc/ldraw-nova.git` and `git clone --branch v0.6.0 https://github.com/anteloc/ldraw-nova-docker.git`. 3. Change into the Docker repository with `cd ldraw-nova-docker`, then run `docker compose build` and `docker compose up -d`. 4. Open http://localhost:8765, or https://localhost:8443 for VR on Meta Quest 3, and accept the self-signed certificate. 5. Stop the app with `docker compose down` when finished. ### Prompt to paste into your agent ```text Clone the ldraw-nova and ldraw-nova-docker repositories at tag v0.6.0 side by side, then run docker compose build and docker compose up -d inside ldraw-nova-docker. Report the local URL and confirm that http://localhost:8765 responds before finishing. ``` ### Caveat The app has no login and the README says to run it only on a trusted network, and the first Docker build needs about 5 GB of disk space. 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/anteloc-ldraw-nova/run-ldraw-nova-web-app-locally/