# Kit USD Agents AI-powered development tools for NVIDIA Omniverse, including Chat USD and Model Context Protocol (MCP) servers for Kit, USD, OmniUI, and Isaac Sim development. ## What's Included This repository provides two types of AI development tools: ### 1. Chat USD Extension Chat USD is an AI assistant that runs inside NVIDIA Omniverse Kit, enabling natural language interaction with USD scenes. **Capabilities:** - **USD Code Generation**: Generate and execute USD Python code from natural language - **Asset Search**: Find USD assets using conversational queries - **Scene Analysis**: Get information about your current USD scene - **Real-time Editing**: Modify scenes interactively through conversation - **Custom Agents**: Extensible architecture for adding specialized capabilities ### 2. MCP Servers Four standalone [Model Context Protocol](https://modelcontextprotocol.io/) servers that integrate with AI coding assistants like Claude, Cursor, and other MCP-compatible tools. #### USD Code MCP Server Provides USD/OpenUSD API assistance with 7 specialized tools: - Browse USD modules and classes - Get detailed class, module, and method API documentation - Search code examples with semantic search - Search USD knowledge base #### Kit MCP Server Comprehensive Kit development assistance with 12 specialized tools: - Search 400+ Kit extensions - Explore extension APIs and dependencies - Find code examples and test patterns - Access Kit documentation and settings #### OmniUI MCP Server OmniUI development assistance with 10 specialized tools: - Browse UI classes, modules, and methods - Get widget documentation, examples, and instructions - Search code examples and window-pattern examples - Look up styling docs and scene-system APIs #### Isaac Sim MCP Server Isaac Sim development assistance with 5 specialized tools: - Search Isaac Sim extensions and APIs - Find code examples with semantic search - Access Isaac Sim documentation and best practices - Search Isaac Sim configuration settings ## Getting Started ### Chat USD Extension Build and run in Omniverse Kit: ```bash # Windows build.bat -r _build\windows-x86_64\release\omni.app.chat_usd.bat # Linux ./build.sh -r ./_build/linux-x86_64/release/omni.app.chat_usd.sh ``` ### MCP Servers Each MCP server can be run locally with Docker or Python. See the individual server documentation: - [USD Code MCP](source/mcp/usd_code_mcp/README.md) - [Kit MCP](source/mcp/kit_mcp/README.md) - [OmniUI MCP](source/mcp/omni_ui_mcp/README.md) - [Isaac Sim MCP](source/mcp/isaacsim_mcp/README.md) **Quick Start (Docker - Recommended):** ```bash # Set your NVIDIA API key first export NVIDIA_API_KEY=REPLACE_WITH_NVIDIA_API_KEY # Linux/macOS # or set NVIDIA_API_KEY=REPLACE_WITH_NVIDIA_API_KEY # Windows # Build wheels and start all MCP servers cd source/mcp ./build-wheels.sh all # or build-wheels.bat all on Windows docker compose -f docker-compose.ngc.yaml up --build ``` This uses NVIDIA's cloud-hosted embedder and reranker services via your API key. No local GPUs required. **Local GPU Deployment (Advanced):** For running embedder/reranker models locally on your own GPUs (requires 2 NVIDIA GPUs): ```bash export NGC_API_KEY=REPLACE_WITH_NGC_API_KEY # For pulling NIM container images export NVIDIA_API_KEY=REPLACE_WITH_NVIDIA_API_KEY cd source/mcp ./build-wheels.sh all docker compose -f docker-compose.local.yaml up --build ``` See [Local Deployment Guide](source/mcp/LOCAL_DEPLOYMENT.md) for details. **Requirements:** - [NVIDIA API Key](https://build.nvidia.com/) (for embeddings, reranking, and LLM access) - Docker (recommended for deployment) ## Documentation - [Chat USD Architecture](source/extensions/omni.ai.chat_usd.bundle/docs/README.md) - [MCP Local Deployment Guide](source/mcp/LOCAL_DEPLOYMENT.md) ## Contributing This project is currently not accepting contributions.