shaderc-vkrunner-mcp

Overview

The shaderc-vkrunner-mcp provides a secure sandbox for AI agents to develop, compile, optimize, and run GPU shaders entirely locally. It uses Mesa's software Vulkan implementation to run shaders on CPU, eliminating risks of GPU crashes or device issues. This approach enables testing advanced features like atomics, subgroups, and raytracing without physical GPU hardware, making it ideal for AI model experimentation with shader code.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• compile_run_shaders

  Compile and run GPU shaders using shaderc and VkRunner

  Try:Compile and run this GLSL shader for meTry:Help me optimize this Vulkan compute shaderTry:Visualize the output of this fragment shader

Note: Tool name was inferred from the VS Code Copilot documentation section that mentions using the '#compile_run_shaders' tag in Agent mode. The README doesn't have an explicit tools section listing all available functions, but suggests shader c

Comparable tools

Installation

git clone https://github.com/mehmetoguzderin/shaderc-vkrunner-mcp
cd shaderc-vkrunner-mcp

docker build -t shaderc-vkrunner-mcp -f Dockerfile .

"shaderc-vkrunner-mcp": {
    "type": "stdio",
    "command": "docker",
    "args": [
        "run",
        "--rm",
        "-i",
        "-v", "${workspaceFolder}:/work",
        "shaderc-vkrunner-mcp",
        "--work-dir",
        "/work"
    ]
}

FAQ

What hardware is required to use this MCP server?

Only Docker is required. The server uses CPU emulation to run shaders, so no physical GPU is needed.

Can I use this with other MCP clients besides VS Code?

Yes, it's compatible with any MCP client. You can test it with MCP Inspector using the provided command.

On Hacker News

Recent discussion from the developer community.

• Show HN: A toy MCP for AI agents to code, run, and see output of GPU code safely▲ 3
  Story by mehmetoguzderin · 2025-04-13

Compare shaderc-vkrunner-mcp with