# Scenario for Blender [Health: Active]

**Category:** 🎥 Multimedia Process  
**Repository:** https://github.com/scenario-labs/blender-plugin  
**GitHub Stars:** 562  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/blender-plugin

## Description
Blender 5.0+ extension for Scenario AI images, video, 3D and PBR materials, with a local MCP server for agents.

## Claude Desktop Quick Installation
No confirmed install details yet, so no config is published. Follow the setup steps in the repository above.

## Documentation

## What Scenario for Blender does

Scenario for Blender MCP server links an authorized agent with the currently open Blender scene and the generation workflow available through the Scenario extension. The local connection is intended for scene-aware agent workflows, while Scenario's hosted MCP service covers platform-wide collections, training, workflows, and usage without requiring Blender.

The Blender extension itself adds creation and review features for image, video, 3D, PBR material, audio, environment, and 360 content. It also includes scene captures, selected-mesh inputs, render workflows, camera path planning, greybox blockouts, a floating composer, generation history, Studio, and film tasks. These capabilities are provided through the Blender extension; the README specifically describes the local MCP server as connecting an agent to the open scene and generation into it.

## How it works

The extension starts a local MCP endpoint at `http://127.0.0.1:9876/mcp`. The Agents (MCP) area reports its status and provides Connect buttons that add the server as `scenario-blender` to supported clients. The local server does not need a separate credential, but the Blender extension still needs authorized Scenario access for generation.

An agent must have an open Blender scene to use the scene connection. For access to Scenario resources that do not depend on Blender, the README recommends using the hosted `mcp.scenario.com` server as well. The two servers serve different scopes: the local endpoint is connected to Blender, while the hosted endpoint provides platform-level access without Blender.

## Setup and configuration

Use Blender 5.0 or later and install the Scenario extension from its remote repository at `https://blender.scenario.com/repo/index.json`, or install a release ZIP manually. Enable Blender's Allow Online Access setting before installation, updates, or communication with Scenario.

A Scenario account with API access, an API key, and its secret are required by the extension. Enter these under the Scenario add-on preferences, test the connection, and choose an output folder. Generated results consume Creative Units, so review the displayed estimate before starting a generation.

After installation, open a Blender scene, open the Scenario panel, and use Agents (MCP) to check the local server and connect a client. The project also documents ZIP verification and separate update behavior for repository and ZIP installations.

## Tools and capabilities

The Scenario for Blender MCP server provides the agent connection to the open scene and Scenario generation. The extension exposes workflows for:

- Text- or reference-based image generation
- Video from prompts, images, or scene playblasts
- Mesh generation and provider edit tasks
- Saved PBR material generation and application
- Speech, music, and sound-effect generation
- Scene-based stills, video renders, and camera planning
- Greybox blockout creation from scene descriptions
- 360 panoramas, HDRIs, and environment upscales
- Generation history, Studio workflows, and film tasks

## Limitations and notes

Scenario for Blender MCP server is marked experimental. It depends on Blender 5.0 or later, an internet connection, Blender's online-access setting, and Scenario account credentials for the extension's API operations. Creative-unit costs vary with models, parameters, and references.

The local MCP endpoint is intended for the open Blender scene and does not replace the hosted Scenario MCP server for platform-wide collections, training, workflows, or usage. Review the project's known limitations before depending on a particular workflow.

_Full upstream README: https://allmcps.com/mcp/blender-plugin/readme_