MCP
Overview
The Model Context Protocol (MCP) is an open standard for connecting AI agents to tools and data. StudioTwin exposes its generation toolkits through MCP-compatible connectors, allowing supported agents to generate assets and work with them in the connected host application.
Agents you already use, including Claude Code, Cursor, VS Code, Gemini and Codex, can connect to StudioTwin through the MCP surface configured for the workflow you choose.
To get started quickly, give your agent the command below to install our StudioTwin MCP Skill:
npx skills add realtwin/studiotwin-mcp-skill
Alternatively, follow the setup guide for your host application.
Setup guides
Choose one ordered path:
- Web / no DCC: create an account and API key → connect StudioTwin Web MCP → use Web MCP.
- Blender: create an account and API key → connect StudioTwin Web MCP → complete Blender-specific setup → use the Blender workflow.
- Unreal Engine: create an account and API key → install the StudioTwin UE plugin → use UE MCP.
StudioTwin Web MCP can operate standalone. Blender depends on it for every StudioTwin cloud operation; the Blender-specific guide does not repeat its endpoint or authentication setup.
Capabilities
The capabilities available to your agent depend on the connector and host you choose. StudioTwin MCP workflows can help you:
- generate assets for environments, materials, meshes, motion, and audio;
- create and refine Concepts, review image variations, and choose a direction for downstream work;
- monitor generation progress and retrieve completed outputs; and
- work with supported results in a connected host application.
Your agent should check the live capabilities when it connects. If you prefer a visual workflow for Concepts, use ConceptLab.