blender-animation-mcp

blender-animation-mcp

Enables multimodal models to drive Blender animation through precise timeline and keyframe tools, import GLB/glTF assets, inspect rigs, and render frames or contact sheets for visual feedback.

Category
Visit Server

README

Blender Animation MCP

An animation-specialized MCP server for Blender. It combines precise timeline/keyframe tools with rendered visual feedback, so a multimodal model can inspect what it animated instead of reasoning from transforms alone.

import_glb_asset loads textured or rigged GLB/glTF assets without flattening their hierarchy; inspect_rig exposes skeletons and bindings; and key_pose_bone provides a safer pose interface than raw RNA paths. Existing embedded animation clips are imported by Blender's glTF importer and included in visual review.

What it can see and control

  • Render any exact frame and return it as MCP image content.
  • Render a contact sheet sampled over the shot for fast visual review.
  • Sample evaluated world transforms and bounding boxes across time.
  • Inspect actions, F-curves, handles, interpolation, NLA strips, constraints, cameras, and markers.
  • Insert/delete keys, change interpolation, control the timeline, and save the file.
  • Import arbitrary .glb and .gltf assets with their hierarchy, materials, rigs, and animation clips.
  • Inspect generated skeletons and key bones directly with local-space quaternions.
  • Discover weighted limb-chain candidates and inspect their actual local axes.
  • Probe bone deformation non-destructively before authoring poses.
  • Create paw IK controls, key them in world space, and validate planted-foot sliding.

The contact sheet is made from actual scene renders. It is deliberately capped at 25 frames; for detailed review, call render_frame on suspicious frames. Animation is temporal, so no finite sample proves every in-between frame is correct—the intended workflow is contact sheet, targeted frames, numeric evaluation, then another visual pass.

Install

  1. Zip the blender_addon directory (the zip must contain blender_addon/__init__.py) and install it from Edit > Preferences > Add-ons > Install from Disk. Enable Animation MCP Bridge.

  2. The bridge starts automatically with Blender. Use Scene Properties > Animation MCP to see its status or stop/restart it.

  3. Install this MCP package:

    pip install -e .
    
  4. Add it to your MCP client configuration:

    {
      "mcpServers": {
        "blender-animation": {
          "command": "blender-animation-mcp",
          "env": {
            "BLENDER_MCP_HOST": "127.0.0.1",
            "BLENDER_MCP_PORT": "9876",
            "BLENDER_MCP_TOKEN": ""
          }
        }
      }
    }
    

For security, the bridge binds to localhost by default. If you configure a token in Blender, set the same BLENDER_MCP_TOKEN for the MCP process. Do not expose the port to an untrusted network.

Recommended model loop

  1. inspect_scene and inspect_object_animation before editing. For a new GLB/glTF asset, call import_glb_asset first and inspect_rig if it is rigged.
  2. Set keys and interpolation.
  3. inspect_animation_visual across the whole shot.

For video-driven pose matching, call setup_video_reference with the source MP4, then use render_video_overlay for exact key poses or inspect_video_overlay for a contact sheet. Adjust video_frame_offset to align the Blender timeline with the useful interval of the reference before keying the rig. For unfamiliar generated rigs, run discover_limb_chains, inspect candidates with inspect_bone_axes, and confirm hinge behavior using probe_bone_deformation. Only then create paw controls with create_limb_ik; animate them through key_paw_pose and check contacts with validate_walk_cycle. 4. Use render_frame around contacts, extremes, and visible discontinuities. 5. Use evaluate_animation when exact spacing or world transforms matter. 6. Iterate, then save_blend.

Environment variables

BLENDER_MCP_HOST, BLENDER_MCP_PORT, BLENDER_MCP_TOKEN, and BLENDER_MCP_TIMEOUT configure the MCP-to-Blender connection.

To run Blender without a window, after enabling the add-on:

blender --background --python blender_addon/headless.py

The included systemd/blender-animation-mcp.service runs this worker persistently on port 9877.

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured