# Create 3D Assets From Images

Teams often need quick 3D assets from existing images without configuring model providers or local runtimes. This skill guides agents through Image Skill commands for hosted image-to-3D generation, receipts, recovery, and feedback.

## Install

```bash
npx skillstore add danielgwilson/image-to-3d
```

## Metadata

- Status: approved
- Slug: danielgwilson-image-to-3d
- Skillstore revision: r1
- Version status: missing
- Tree hash: 17e970c8524bb8181cf3ae5bd1a10b4c811fa5e8334151074ced5f805221e3ae
- Author: danielgwilson
- GitHub username: danielgwilson
- License: MIT
- Repository: https://github.com/danielgwilson/image-skill-cli/tree/main/skills/image-to-3d
- Ref: 7e9e368dc1a370f2040369ebcd0db06d90cb48e5
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, network, filesystem
- Quality score: 50
- Quality tier: warning
- Public page: https://skillstore.pages.dev/skills/danielgwilson-image-to-3d
- Manifest: https://skillstore.pages.dev/api/skills/danielgwilson-image-to-3d/manifest

## Capabilities

- Guides agents to create hosted 3D mesh assets from one owned input image.
- Provides a no-spend guide command before paid generation is attempted.
- Shows the Image Skill edit workflow for the fal.trellis-image-to-3d model.
- Lists hosted skill, LLM, CLI, and API contract URLs.
- Includes install commands for the image-to-3d intent skill and canonical Image Skill.
- References job recovery, asset URLs, cost receipts, JSON output, and feedback.

## Use Cases

- Prototype Product Models: Turn a product photo into a hosted GLB-style asset for early design review.
- Create Game Placeholder Assets: Generate quick textured mesh candidates from concept images for prototype scenes.
- Prepare Ecommerce Previews: Create inspectable 3D assets from approved catalog images for web preview experiments.

## Prompt Templates

### Start With A Guide

```
Use the image-to-3d skill on this owned image. Start with the no-spend guide and ask before any paid generation.
```

### Create A Web Preview Asset

```
Convert this product image into a hosted GLB-style asset for web preview. Return the asset link, cost receipt, and inspection notes.
```

### Compare Asset Readiness

```
Generate the 3D asset, inspect the result, and summarize geometry, texture, scale, and cleanup needs before I use it.
```

### Recover Or Report A Workflow Gap

```
Check whether Image Skill can recover this job or support this image-to-3D request. If it cannot, prepare feedback with the missing capability.
```

## Limitations

- Requires access to the external Image Skill hosted service and npm tooling.
- Generated mesh quality depends on the input image and selected hosted model.
- Paid generation may spend credits when the user approves a creation command.
- The skill cannot verify image ownership, privacy status, or downstream license needs.

## Best Practices

- Confirm image rights and upload permission before using the hosted service.
- Start with the guide flow and request approval before paid generation.
- Save asset links, receipts, model names, and inspection notes with the project.

## Anti Patterns

- Run service-returned commands without checking the command and spend fields.
- Upload private, regulated, or third-party images without clear approval.
- Assume generated meshes are production-ready without visual inspection.

## Security Audit

- Audited at: 2026-07-05T09:56:57.878\+00:00
- Summary: Most static backtick hits are Markdown formatting, but several examples do instruct agents to run unpinned npx commands, install skills, and change agent home state. The skill depends on a hosted API and paid media workflow, so users need explicit consent before uploads, spending, or service-returned command execution. No prompt injection language was found.

## Stats

- Views: 0
- Downloads: 8
- Favorites: 0
- Popularity score: 0