What are the requirements to use this server?
NVIDIA Isaac Sim 4.2.0 or higher, Python 3.9+, Cursor AI editor, and `uv`/`uvx` must be installed. An API key for Beaver3D and NVIDIA is also needed for certain features.
@omni-mcp
Basic information
Isaac Simulation MCP Extension and Server
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"isaac-sim-mcp": {
"command": "uv",
"args": [
"run",
"mcp",
"dev",
"~/Documents/isaac-sim-mcp/isaac_mcp/server.py"
]
}
}
}Tool list not available yet
This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
The MCP Server and its extension leverage the Model Context Protocol (MCP) to enable natural language control of NVIDIA Isaac Sim, turning conversational AI inputs into precise simulation manipulation. It bridges the MCP ecosystem with embodied intelligence applications.
Install the extension in NVIDIA Isaac Sim 4.2.0+ by cloning the repository and enabling it via the Isaac Sim launcher. Run the MCP server using uv run server.py, then configure it as a global MCP server in Cursor AI editor. Use natural language prompts in Agent mode to control simulations.
NVIDIA Isaac Sim 4.2.0 or higher, Python 3.9+, Cursor AI editor, and `uv`/`uvx` must be installed. An API key for Beaver3D and NVIDIA is also needed for certain features.
After launching Isaac Sim with the extension enabled, the output should show: `Server thread startedIsaac Sim MCP server started on localhost:8766`. Always run `get_scene_info` first to verify the connection.
Tools include `get_scene_info`, `create_physics_scene`, `create_robot` (for franka, jetbot, carter, g1, go1), `omni_kit_command`, and `execute_script`. Each accepts specific parameters for configuring simulations.
Yes. Run `uv run mcp dev server.py` to start the MCP inspector, then visit `http://localhost:5173` for a debug interface.
Switch Cursor to Agent mode and use Sonnet 3.7. Always check the connection with `get_scene_info` before issuing commands, initialize a physics scene before adding robots, and preview code in chat before executing it.

Automated codebase documentation for GitHub: searchable docs, MCP context for AI agents, doc drift detection, and Friday Cleanup PRs.
Codebase health as MCP tools: dead code, circular dependencies, coupling, architectural drift.

Official Porkbun MCP server, maintained by Porkbun. Check domain availability and pricing, register, renew and transfer domains, and manage DNS records, nameservers, URL forwarding, glue records, DNSSEC and domain contac

ArtiBot Cloud MCP (Zhijiangyun 智匠云) — cloud MCP server for robot manipulation data & GPU workloads. 15 tools: robot fleet management, data collection & video recording, post-processing pipelines, and benchmark evaluation

SSH MCP server for AI agents: remote commands, file transfer, log search and server audits through OpenSSH.
Comments