Skip to main content
Glama
whats2000

Isaac Sim MCP Server

by whats2000
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Related Servers

Alternatives to Isaac Sim MCP Server

  • omni-kit-mcp

    inbarajaldrin
    A
    license
    A
    quality
    B
    maintenance
    Enables driving Omniverse Kit apps (Isaac Sim, Isaac Lab) over MCP, allowing agents to control simulations, run Python, and call namespace-scoped tools via a single bridge.
    1
    1
    MIT

  • IsaacSim-MCP

    Tim0320
    A
    license
    C
    quality
    B
    maintenance
    Enables AI assistants to control NVIDIA Isaac Sim by building scenes, loading assets, operating robots and humans, reading sensors, managing simulation, and creating Action Graphs.
    129
    1
    MIT

Related Servers

  • nvidia-isaac-mcp

    nullbyte91
    A
    license
    Not graded
    quality
    D
    maintenance
    Enables AI tools like Windsurf and Claude to control NVIDIA Isaac Sim and Isaac Lab through natural language, providing tools for scene inspection, prim management, physics simulation, and robot spawning.
    8
    Apache 2.0
  • simliverse-mcp by SimLiverse

    simliverse-mcpofficial

    SimLiverse
    A
    license
    Not graded
    quality
    B
    maintenance
    Provides a unified MCP server for the SimLiverse Physical AI Platform, enabling LLM copilots to access Isaac Sim knowledge via NVIDIA NIM and execute/control simulation through a TCP socket.
    MIT

  • MCP Unity Editor

    AboJad2026
    A
    license
    Not graded
    quality
    Not graded
    maintenance
    Enables AI assistants to interact with Unity Editor through the Model Context Protocol, allowing natural language control of Unity projects including scene manipulation, GameObject creation, component updates, package management, and test execution.
    -
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.7/5.0

Scored across 42 tools

Disambiguation5/5

Every tool targets a distinct asset, operation, or domain (simulation control, scene creation, sensors, materials, action graphs, diagnostics). Even superficially similar tools like get_physics_state, get_joint_config, and get_robot_info are clearly separated by their descriptions and use cases. No two tools appear to do the same thing.

Naming Consistency5/5

All 42 tools follow a consistent verb_noun snake_case pattern: play_simulation, create_object, get_robot_info, list_environments, delete_object, etc. Verbs are clear and predictable ('create', 'get', 'list', 'set', 'load', 'delete'), with no mixed conventions or vague names like 'process' or 'do_thing'.

Tool Count2/5

42 tools far exceeds the 3-15 well-scoped range and even the 25+ threshold classified as 'too many'. While the broad simulation domain justifies some breadth, this surface is heavy and could overwhelm an agent with selection overhead. It sits between borderline and extreme, so a 2 is appropriate.

Completeness4/5

The tool set covers the full lifecycle for core entities: objects (create, read, transform, delete), robots (create, inspect, command, configure), environments (load, list, clear), materials (create, apply), lights (create, modify), sensors (create, capture, read), and simulation control. A minor gap is lack of direct joint velocity reads, but execute_script and step_simulation observations partially fill it, so no dead ends.

Maintenance

ActivityActive
ResponsivenessResponsive

Resources