BlenderMCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| BLENDER_HOST | No | Host address for Blender socket server (default: "localhost") | localhost |
| BLENDER_PORT | No | Port number for Blender socket server (default: 9876) | 9876 |
| BLENDER_AUTH_TOKEN | No | Optional shared secret. If set in Blender (Addon Preferences) the MCP server must send the same token. | |
| BLENDER_MCP_TOOL_PROFILE | No | Controls how many MCP tools are exposed (default: "standard"). Profiles: minimal, standard, full. | standard |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {
"tasks": {
"list": {},
"cancel": {},
"requests": {
"tools": {
"call": {}
},
"prompts": {
"get": {}
},
"resources": {
"read": {}
}
}
}
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| assign_materialC | Assign a material to a mesh object slot. |
| create_primitiveC | Create a primitive mesh object. |
| execute_blender_codeB | Execute arbitrary Python code in Blender. Make sure to do it step-by-step by breaking it into smaller chunks. Parameters:
|
| execute_opsC | Execute a validated allowlisted ops request (DSL v1) in Blender. Parameters:
|
| get_capabilitiesA | Return server/addon capabilities: DSL version, enabled integrations, supported ops/formats. |
| get_scene_infoB | Get detailed information about the current Blender scene |
| get_viewport_screenshotA | Capture a screenshot of the current Blender 3D viewport. Parameters:
Returns the screenshot as an Image. |
| render_stillC | Render a still image to output_path (or temp). |
| select_objectsC | Select objects by name.
|
| set_transformC | Set object transform values. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| asset_creation_strategy | Defines the preferred strategy for creating assets in Blender |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| help_index | Start here: how to set up and use BlenderMCP. |
| help_quickstart | Minimal steps to connect Blender + MCP client and run a first command. |
| help_tools | High-signal tips for using BlenderMCP tools effectively. |
| help_troubleshooting | Common setup and runtime issues and what to try next. |
| help_security | Key safety considerations when using BlenderMCP. |
| example_ops_basic | Create a cube, make a material, assign it (DSL v1). |
| example_ops_modifiers | Add and apply a bevel modifier (DSL v1). |
| example_workflow_asset_strategy | Recommended step-by-step workflow when creating assets/scenes. |
TDQS
Scored across 10 tools
Most tools have distinct purposes targeting different Blender operations like material assignment, primitive creation, code execution, scene info retrieval, and rendering. However, 'execute_blender_code' and 'execute_ops' could cause confusion as both involve executing operations in Blender, though their parameter structures differ.
The naming follows a consistent verb_noun pattern throughout (e.g., assign_material, create_primitive, get_scene_info), which is clear and predictable. There are no deviations in style, making it easy to understand each tool's function at a glance.
With 10 tools, the count is well-scoped for a Blender automation server, covering key operations like object manipulation, scene management, rendering, and code execution. Each tool appears to earn its place without feeling excessive or insufficient for the domain.
The toolset covers many core Blender workflows, such as object creation, selection, transformation, material assignment, rendering, and scene inspection. However, there are notable gaps, such as the absence of tools for deleting objects, modifying materials, or handling animations, which could limit agents in more complex tasks.