Skip to content
MCP server

DCC-MCP Core

One gateway that lets AI agents operate Maya, Blender, Houdini and dozens of other apps

Install
Pricing
Open source
Platforms
CLI, API, Windows
Runs
Remotely
Version
0.20.51
Listing updated
2 days ago

About DCC-MCP Core

DCC-MCP Core is an open-source MCP server for design and media. It runs as a hosted remote server, so there is nothing to install locally.

dcc-mcp-core is the shared layer under a whole family of host adapters: the README counts 48 adapter entries in its catalog covering desktop DCCs (Maya, Blender, Houdini, 3ds Max, Cinema 4D), engines (Unreal, Unity, Godot), image and video tools (Photoshop, After Effects, Premiere, ComfyUI), CAD tools (FreeCAD, OpenSCAD) and production tracking. It notes that the count includes experimental entries and does not prove release-qualified support for each.

A machine-wide gateway, on port 9765 by default, fronts every live backend behind one Streamable HTTP endpoint. Adapters share one tool contract, an asset-import contract and a write-then-read-back postcondition so a tool proves its own result. A recipe runtime exposes recipes__list, recipes__get, recipes__validate and recipes__apply, driven by declarative RECIPES.yaml files, and a marketplace of skills lists example prompts for each entry.

The core is Rust-powered with Python packaging (pip install dcc-mcp-core), a CLI and a REST interface alongside MCP. The project site shows worked examples with notes on what was and was not verified, such as a Blender render and a Houdini vessel built through typed calls. It is aimed at studios and tool builders wiring agents into creative pipelines.

Install

Run in a terminal

claude mcp add dcc-mcp-core -- uvx dcc-mcp-server

Add to claude_desktop_config.json

{
  "mcpServers": {
    "dcc-mcp-core": {
      "command": "uvx",
      "args": [
        "dcc-mcp-server"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "dcc-mcp-core": {
      "command": "uvx",
      "args": [
        "dcc-mcp-server"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "dcc-mcp-core": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "dcc-mcp-server"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "dcc-mcp-core": {
      "command": "uvx",
      "args": [
        "dcc-mcp-server"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "dcc-mcp-core": {
      "command": "uvx",
      "args": [
        "dcc-mcp-server"
      ]
    }
  }
}

Configuration

  • DCC_MCP_GATEWAY_PORTPort the machine-wide gateway binds. One agent endpoint fronts every live DCC backend, so the port in the transport URL has to match this value.

Tools this server provides (4)

Taken from the project documentation; the running server may differ.

recipes__list
List available declarative recipes.
recipes__get
Fetch one recipe.
recipes__validate
Validate a recipe before use.
recipes__apply
Apply a recipe to a host.

What it does

Single MCP gateway
One endpoint fronts every running creative-app backend on the machine.
Shared tool contract
Every host adapter implements the same protocol and result structure.
Recipe runtime
Hosts ship capabilities as declarative data via list, get, validate and apply.
Skill marketplace
Entries carry example prompts and install metadata.
MCP, CLI and REST
The same capabilities are reachable from agents, scripts and HTTP.
Postcondition checks
Tools write, then read back, to confirm their own results.

Good for

  • Have an agent build and render a material in a live Blender session
  • Script repeatable asset imports across several DCC tools
  • Give studio pipeline tools a single agent-facing endpoint
  • Run traceable creative workflows with logged requests

Pricing

Open source

MIT licensed

Prices change. Check dcc-mcp.github.io before you buy.

DCC-MCP Core FAQ

Which applications are supported?

The README reports 48 adapter entries, with Maya, Blender, Houdini, Unreal, Unity, Photoshop and FreeCAD among examples; some are experimental.

How do I install it?

Run `pip install dcc-mcp-core`; the registry also lists the `dcc-mcp-server` package.

Which licence is it under?

The README shows an MIT licence badge.

What interfaces does it offer besides MCP?

A CLI and REST interfaces, per the project site.

Is DCC-MCP Core free?

Yes. DCC-MCP Core is open source, so you can use and self-host it for free.

How do I add DCC-MCP Core to Claude Code?

Run this command in your terminal: claude mcp add dcc-mcp-core -- uvx dcc-mcp-server

What are the best alternatives to DCC-MCP Core?

The closest alternatives to DCC-MCP Core are REA, HyperFrames by HeyGen, Codebase Memory, GitHub and Serena MCP: the IDE for your agent.

Alternatives to DCC-MCP Core

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Hosted MCP connector for creating and rendering videos from HTML through AI chat

    • Remote
    • Freemium
    • 85.3K GitHub stars
    • Design and media
  3. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  4. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  5. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  6. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory