mcpField
C
Cubric Studio

Cubric Studio

Local
cubric.studioMedia & Design2Updated 1w ago

About Cubric Studio

Cubric Studio connects AI agents to the Cubric Studio desktop app to create and edit images, video, and GIFs. It is for users with Cubric Studio 2.0 or newer open, and saves generated work as gallery cards with prompts and settings.

Key features

  1. 01Generate and edit images and video
  2. 02Create and edit GIFs
  3. 03Create and open projects
  4. 04Run Flows and inspect results
  5. 05Use personal files as references
  6. 06Confirm paid cloud model charges

Use cases

  1. 01Generate an image in a Cubric Studio project
  2. 02Edit video using personal reference files
  3. 03Review generated results in the gallery

MCP support

The MCP server lets compatible agents use Cubric Studio through the app running on the same computer. It provides tools for projects, media generation and editing, Flows, and gallery cards.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository
Packages
mcpbcubric-studio.mcpb0.3.2

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Cubric Studio for AI agents

Let your AI agent make images, video and GIFs in Cubric Studio, the desktop app on your computer. Ask Claude, Codex or Antigravity for "an image of a red bicycle in a new project" and it lands in your gallery as a real card, with its prompt and settings saved.

You need Cubric Studio 2.0 or newer, installed and open. Earlier versions have no agent connection. Your agent reaches the app at http://127.0.0.1:3000/mcp, on your own computer only; in Claude Code, the plugin's Cubric Studio address setting changes it if you started the app on another port.

Install

The easy way: in Cubric Studio, open Settings > Connect an agent and press Connect next to your agent. The app runs the steps below for you. The manual steps work on Windows, macOS and Linux.

Claude Code

/plugin marketplace add https://github.com/MadPonyInteractive/cubric-studio-agents.git
/plugin install cubric-studio@cubric-studio

Use the full https:// address: the short MadPonyInteractive/cubric-studio-agents form clones over SSH, which fails unless you have a GitHub SSH key set up.

Codex

codex plugin marketplace add MadPonyInteractive/cubric-studio-agents
codex plugin add cubric-studio@cubric-studio

Claude Desktop (Windows and macOS; there is no Claude Desktop for Linux): download cubric-studio.mcpb, then in Claude Desktop open Settings > Extensions and drag the file onto that page. Claude shows what it installs and asks you to confirm. Double-clicking the file only works where your computer already opens .mcpb files with Claude.

Antigravity: download this repository (Code > Download ZIP), copy the plugins/cubric-studio folder into %USERPROFILE%\.gemini\config\plugins\ (on macOS and Linux, ~/.gemini/config/plugins/), then restart Antigravity.

ChatGPT: use Codex, which comes with your ChatGPT plan. The ChatGPT chat window runs in the cloud and cannot reach apps on your computer.

Gemini: use Antigravity, Google's agent app for your computer (install steps above). It is free to start with the Google account you use for Gemini, no subscription needed; a Google AI Pro or Ultra plan gives it more use. The Gemini app and gemini.google.com run in the cloud and cannot reach apps on your computer.

What your agent can do

List models and read their prompting guides, create and open projects, generate and edit images and video (with your own files as references), run Flows, look at the results, rename cards, and make and edit GIFs.

  • Paid cloud models ask first. The agent is told the price and must confirm it before anything is billed. Local models are free and never ask.
  • Nothing deletes. No tool removes a project, a card or a file, and none installs a model.

Privacy

This plugin connects your agent to the app on your own computer and nowhere else. Your agent's provider (Anthropic, OpenAI or Google) sees what your agent sees, including the pictures it looks at. What the app itself sends, and where, is listed at cubric.studio/privacy.

Contact

[email protected]

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.