modly-image-to-3d — Claude Code skill: 3D from photos, local
This skill walks your agent through installing Modly, a desktop app that turns a photo into a 3D mesh on your own GPU — no cloud, no API keys.
npx skills add reason-machines/trending-skills --skill modly-image-to-3d -a claude-codeWhat modly-image-to-3d does
Modly is an open-source desktop app for Windows and Linux. Feed it an image and it produces a 3D mesh, running the model entirely on your GPU. The skill covers setup, the dev build and the extension system.
Under the hood it's an Electron and TypeScript shell over a local Python FastAPI backend. Generation lives in each extension's generator.py, while manifest.json describes the model variants. Extensions install straight from a GitHub repo. For a quick start there's launcher.bat on Windows and launcher.sh on Linux; the dev path needs npm install, a Python virtual environment, pip install -r requirements.txt and then npm run dev.
Your GPU is the real gate here. Each variant in the manifest carries a vram_gb figure, and a card below it won't finish the job. Check that number before pulling weights that run several gigabytes — otherwise you wait through a download for nothing.
What it helps with
- turn a single photo into a 3D mesh
- run 3D generation on your own GPU without a cloud service
- install a Modly extension from a GitHub repo
- pick a model variant that fits your VRAM
- start the local FastAPI backend for generation
Example prompts for the agent
- install the modly-image-to-3d skill and show me how to launch Modly
- how do I install a Modly extension from GitHub and download a model variant
- check whether my VRAM is enough for the variant in manifest.json
How to install modly-image-to-3d
1Skills CLI (Vercel)
npx skills add reason-machines/trending-skills --skill modly-image-to-3d -a claude-code2Manually: put the files in place
~/.claude/skills/modly-image-to-3d/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/modly-image-to-3d/
1Skills CLI (Vercel)
npx skills add reason-machines/trending-skills --skill modly-image-to-3d -a codex2Manually: put the files in place
.agents/skills/modly-image-to-3d/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add reason-machines/trending-skills --skill modly-image-to-3d -a cursor2Manually: put the files in place
.cursor/skills/modly-image-to-3d/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install reason-machines/trending-skills modly-image-to-3d2Manually: put the files in place
.github/skills/modly-image-to-3d/SKILL.md1Skills CLI (Vercel)
npx skills add reason-machines/trending-skills --skill modly-image-to-3d -a gemini-cli2Platform CLI
gemini skills install https://github.com/reason-machines/trending-skills --path skills/modly-image-to-3dManually: put the files in place
.devin/skills/modly-image-to-3d/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add reason-machines/trending-skills --skill modly-image-to-3d -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about modly-image-to-3d
How do I install modly-image-to-3d in Claude?
Run npx skills add reason-machines/trending-skills --skill modly-image-to-3d -a claude-code. The skill file then lands at ~/.claude/skills/modly-image-to-3d/SKILL.md.
What does modly-image-to-3d do?
It guides your agent through installing and configuring Modly, the desktop app that generates a 3D model from an image on a local GPU. It also covers adding model extensions and starting the backend.
Do I need API keys or an internet connection to generate?
No API keys — the model runs on your graphics card. You only need the network to download model weights and the extension itself from GitHub.