---
id: dcc-mcp-server
version: "0.20.6"
license: MIT
license_treatment: permissive
maintenance: active
---

# dcc-mcp-server — Standalone dcc-mcp-server CLI binary — gateway, sidecar, stdio bridge (Rust)

License: permissive · Maintenance: active · Downloads: 128.2K/mo

## What it is and what it does

dcc-mcp-server is a Rust binary packaged as a Python wheel that provides a standalone gateway and sidecar process for connecting DCC applications to the Model Context Protocol. Rather than being loaded as a Python extension module into the host DCC interpreter, it runs as its own OS process and communicates with DCC plugins via RPC, making it suitable for operators and sidecar spawners who need a separate, independently-versioned binary.

The package follows the pattern of tools like ruff and uv: a single `pip install` command places a CLI binary on PATH that can be invoked from DCC plugins to establish the gateway connection. It supports Python 3.7 through 3.14 and is distributed as platform-specific wheels for macOS (x86_64 and arm64), Linux (x86_64), and Windows (x86_64 and arm64), with no runtime Python dependencies.

Use it for:

- Spawn a Maya sidecar gateway from a Maya plugin to enable MCP-based tool integration without modifying the host interpreter.
- Deploy a Blender addon that launches the sidecar as a subprocess to bridge Blender's Python with external MCP servers.
- Run the gateway on a headless Linux server to provide an MCP endpoint for Houdini batch processing or render farm jobs.
- Integrate the binary into a DCC plugin installer so end users get a working gateway with a single pip command.
- Use the CLI directly from shell scripts or CI/CD pipelines to test MCP protocol interactions with DCC workflows.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Distributes a Rust-compiled CLI binary that runs as a standalone gateway or sidecar process to bridge DCC applications (Maya, Blender, Houdini) with the Model Context Protocol.

Yes, if you are building DCC plugins or operators that need to integrate with the Model Context Protocol. The package is actively maintained, has no security vulnerabilities, carries a permissive MIT license, and solves the specific problem of distributing a standalone binary gateway without coupling it to the core Python extension module. Install it when you need the sidecar spawner pattern; skip it if you are not working with DCCs or MCP.

## Install

pip install dcc-mcp-server
uv add dcc-mcp-server
poetry add dcc-mcp-server

## Installing dcc-mcp-server

Before you install: Released 1 day ago with active maintenance; installs via platform-specific wheels (macOS, Linux, Windows) with no runtime dependencies, though medium friction reflects the binary-only distribution model and requirement to spawn as a separate OS process.

License in practice: MIT license permits commercial and private use with minimal restrictions; suitable for integration into proprietary DCC plugins and workflows.

Quickstart:

pip install dcc-mcp-server

import subprocess
from dcc_mcp_server import binary_path

proc = subprocess.Popen([
    str(binary_path()),
    "sidecar",
    "--dcc", "maya",
    "--host-rpc", "commandport://127.0.0.1:6000",
])

Requires Python 3.7 or later; binary must run as a separate OS process, not loaded into the host Python interpreter.

Verify before relying:

- Exact command-line interface and all available sidecar flags beyond the example shown.
- Whether the binary works on embedded Python interpreters (e.g. Maya 2022's Python 3.7) in practice.
- Performance and resource overhead when running as a long-lived sidecar process.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 128.2K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

dcc gateway sidecar binary, maya blender houdini mcp bridge, standalone mcp server cli, dcc plugin sidecar spawner, model context protocol gateway, dcc-integration, mcp-gateway, rust-binary

[View on SkillFed](https://skillfed.io/packages/dcc-mcp-server) · [View on PyPI](https://pypi.org/project/dcc-mcp-server/)