1. Conduid
  2. Developer Tools
  3. Openusd AI
MCP server · Developer Tools

Openusd AI

A collection of USD tools that use a number of AI interfaces MCP and Agent2Agent.

Unclaimed MIT last commit a year ago devtools
56Fair

Scored 5 months ago · breakdown

About Openusd AI

Openusd AI is an MCP server published by rudycazabon in the Developer Tools category: a collection of USD tools that use a number of AI interfaces MCP and Agent2Agent.

The repository has 19 stars and 3 forks, with the last commit a year ago. Six months or more without a commit doesn't mean the server is broken, but check the open issues before depending on it in production.

Install

From source
git clone https://github.com/rudycazabon/OpenUSD-AI

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Openusd AI

Powered by Claude · Grounded in docs

I know everything about Openusd AI. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

Static checks derived from the repository 5 days ago. Heuristics on the source, not an audit.

  • ✓README presentThe repository ships documentation.
  • ✓License declaredUsable under a stated license.
  • ✓Tests presentA test suite exists in the repository.
  • ✓Dependencies pinnedA lockfile pins dependency versions.
  • ✓No dynamic code executionSource scan found no eval, dynamic Function, or shell execution.
  • !Scoped permissionsDoesn't declare a permission scope. Assume it can do anything its process can.

README

USD AI

The contents of this repository are a collection of OpenUSD tools that interface with MCP (modelcontextprotocol) and A2A (agent2agent). The goal of this effort is to give large language models and agentic workflows the tools to work directly with OpenUSD objects, e.g., UsdStage, prims, hierarchies, and others and integrate them into comprehensive asset creation workflows.

Introduction

Setup

Currently this project has been tested to work on Windows 11 with usd-core from PyPi and NVidia USD v25.05 with Python 3.11.11.

  • Install UV package manager from https://astral.sh
  • Execute the following:
  • To test out the server
    • python -m pytest usd-mcp/server/tests/test_server.py
    • python -m pytest usd-mcp/server/tests/test_usd_mcp_server.py
  • Adding to the Claude Desktop (Windows)
    • Open %APPDATA%\claude_desktop_config.json
    • Add/Modify the file with the following:
{
  "mcpServers": {
    "usd-mcp": {
      "command": "<FULL PATH>\\usd-ai\\.venv\\Scripts\\python.exe",
      "args": ["<FULL PATH>\\usd-ai\\usd-mcp\\usd_mcp\\main.py"],
      "cwd": "<FULL PATH>\\usd-ai"
    }
  }
}

Usage

Within the context for Claude Desktop (Windows) you will have access to the following USD tools:

  • load_usd_stage
  • get_stage_hierarchy
  • inspect_prim
  • list_stage_prims
  • find_prims_by_name

For the purposes of this repo I have included usd-mcp\data\HelloWorld.usda. I have loaded the Pixar KitchenSet, however, your milage will vary (trust me).

A sample prompt session that has worked"

Load the USD file at <FULL PATH>\\usd-ai\\usd-mcp\\data\\HelloWorld.usda.
<will see load of file>

Show me the hierarchy and present in formatted JSON.

Contents

usd-mcp - OpenUSD tools exposed with MCP protocols.

usd-mcp README.md

usd-a2a - OpenUSD tools exposed to agentic workflows with the a2a protocol.

README fetched from the repository 5 days ago. The original is authoritative.

Questions

About Openusd AI

How do I install Openusd AI?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Openusd AI safe to use with an AI agent?

Its trust score is 56 out of 100 (fair). It passes 5 of the 6 static security checks run so far; the failures are listed above. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Openusd AI still maintained?

The last commit was a year ago. That's long enough that you should check whether the maintainer is responding to issues before depending on it.