I Gave AI Access to Blender. Here's What It Built

Hello, I'm Rijul, and I'm building LiveReview — a blast-radius aware AI code review built for your business-critical systems. Star us to help devs discover the project, give it a try, and share your feedback to help improve the product.

We see MCPs being used for a wide variety of tasks.

But have you imagined how they can help with software like Blender?

Let's take a look.

This is the repository:

https://github.com/ahujasid/mcp-for-blender

This is a plugin that lets you control Blender with any LLM of your choice. You can prompt the LLM, and Blender can act accordingly.

Which is pretty cool.

To install it, you can follow the instructions in the README.

Step 1: Install UV

First, you need to install UV.

brew install uv

curl -LsSf https://astral.sh/uv/install.sh | sh

powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

Step 2: Add the MCP server to your client

Claude Desktop:

{
    "mcpServers": {
        "blender": {
            "command": "uvx",
            "args": ["mcp-for-blender"]
        }
    }
}

Claude Code:

claude mcp add blender uvx mcp-for-blender

Codex:

codex mcp add blender -- uvx mcp-for-blender

Or, if you are using OpenCode, use this:

{
  "mcp": {
    "blender-mcp": {
      "type": "local",
      "command": ["uvx", "mcp-for-blender"],
      "enabled": true,
      "environment": {
        "BLENDER_HOST": "localhost",
        "BLENDER_PORT": "9876"
      }
    }
  }
}

Step 3: Install the Blender add-on

uvx mcp-for-blender install-addon

Step 4: Go to Blender → Edit → Preferences

Step 5: Click Add-ons and check MCP for Blender

Step 6: Connect the MCP server

In Blender, press N. The following will appear on the screen.

Now click the MCP for Blender tab.

Ensure that the server is connected.

Under Asset Libraries, check Poly Haven.

Poly Haven is a free library of high-quality HDRIs, textures, and 3D models that anyone can use commercially without attribution.

The setup is done. Let's get to the fun part.

This is my current Blender scene:

It's the default scene, with the cube, light, and camera.

Let's try asking the MCP about this scene.

Open your agent. Here, I'm using OpenCode.

What's currently in my Blender scene?

As you can see, it accurately describes the current scene.

Now let's put it to more work, I'll ask it to make a simple room.

Delete the default cube. Create a simple room: a floor and four walls, about 5x5 meters, 3 meters tall.

We got ourselves a room.

So, this was a basic example. Now let's make a scene.

I'm trying out this prompt, this creates an island with textures and lighting from Poly Haven.

Build a striking low-poly floating island scene in Blender. Steps:

1. Clear the default scene.
2. Create a floating island base using a deformed icosphere or cube with noise-based displacement, flattened on top for a grassy surface and jagged on the underside for rock.
3. Add a simple low-poly tree on top, slightly off-center.
4. Scatter 4-6 small rocks on the surface at varied scale and rotation.
5. Apply Poly Haven textures: grass/moss on top, rock/stone on the underside.
6. Add a Poly Haven HDRI for a dramatic sunset or golden-hour sky.
7. Position the camera for an epic three-quarter low-angle shot that makes the island feel large and floating.
8. Render the final image.

Use Poly Haven assets for all textures and lighting. Favor bold silhouette and dramatic lighting over fine geometric detail.

Once it has finished building the scene, I get the following:

When I render it, I get the following output:

You can prompt it again and make all the tweaks you need until you are satisfied :)

You don't need to be a Blender pro to quickly create the 3D models or scenes you need.

With MCP, you can describe what you want in natural language, and the agent can interact with Blender to create and modify the scene for you.

You can start with something as simple as asking what's currently in the scene, then move on to creating a room, and eventually build a much more complex scene with a single prompt.

Your team's attention is limited, and the deluge of AI-generated code is making it harder to keep production reliable and secure without slowing you down.

I'm building LiveReview, a blast-radius aware AI code review built for your business-critical systems.

Instead of presenting every diff with equal emphasis, LiveReview scores each change by blast radius — how far its impact reaches through your call graph — so you can focus attention where it actually matters.

Spend code review effort where business risk is highest — not spread evenly across every diff.

⭐ Star it on GitHub:

LiveReview is an AI code reviewer that scores every hunk of a diff by blast radius: how far a change reaches through your call graph, how much persistent state it touches, and how well-tested it is. A 3-line change to a shared auth check can outrank a 300-line UI tweak. Your team's attention goes to the highest-risk code first, not spread evenly across every diff.

LiveReview's Blast Radius & Review Priority scoring, live in the diff viewer.

The exact math, not a black box	Visualize blast radius at a glance	Every factor that feeds the score

Here's the goal:

Click below to try LiveReview with your codebase:

source & further reading

dev.to — original article Beyond Generative LLMs: Architecting Sub-10ms Voice AI with Felona Voice We configured residential proxies on our MCP server. They did nothing. Six releases later. Jev: The Frontier Model That Refuses to Write