bonsai-mcp

Create and inspect IFC4 building models with IfcOpenShell in the Bonsai MCP workspace.

465|41|Updated Aug 4, 2026
One-click install
npx skills add https://github.com/autonomous-ai/openharness --skill bonsai-mcp-autonomous-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bonsai-mcp
Source: https://github.com/autonomous-ai/openharness/tree/main/store/agents/bonsai-mcp/skills/bonsai-mcp
Command: npx skills add https://github.com/autonomous-ai/openharness --skill bonsai-mcp-autonomous-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires ifcopenshell.

What problem does it solve? Authoring and verifying BIM building data programmatically is difficult: IFC models are complex, and it is hard to confirm that generated geometry, quantities, and spatial structure are correct. This Skill provides a controlled workflow to build real IFC4 buildings, inspect their contents, and verify quantities against measured outputs. ## Core Features & Use Cases - IFC4 Building Authoring: Configure footprint, storeys, height, and room use, then run a build that produces a real IFC4 model with storeys, slabs, walls, spaces, and quantity sets using IfcOpenShell. - Model Inspection and Verification: Reopen the generated building.ifc, inspect spaces, walls, and openings, and match areas and volumes against quantities.csv. - Optional Blender/Bonsai Bridge: Connect to an existing Blender session through the pinned upstream integration for native Bonsai editing and read-back verification. - Use Case: An architect wants to prototype a three-storey office layout, generate the IFC model, and confirm that slab areas and space volumes match the intended program before handing the file to downstream BIM tools. ## Quick Start Ask the agent to build a two-storey office building with a 20 by 15 meter footprint and verify the resulting IFC spaces and quantities.

Frequently Asked Questions about bonsai-mcp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create an IFC building model with IfcOpenShell?

Configure the footprint, storeys, height, and room use in the studio controls, then run the toolchain build script. The workflow generates a real IFC4 file with storeys, slabs, walls, spaces, and quantity sets using IfcOpenShell.

How do I verify quantities in a generated IFC model?

Reopen building.ifc with IfcOpenShell, inspect the spaces, walls, and openings, and read the quantity sets. Match their areas and volumes against the quantities.csv produced by the run to confirm correctness.

Can I use this with Blender and Bonsai for BIM editing?

Yes, an optional pinned bridge connects to an existing Blender session for native Bonsai editing. The wrapper's bonsai action only requests scene information from the running session; read the upstream docs before using native commands.

Does the in-pane drawing certify structural or code compliance?

No. The in-pane drawing is a schematic model inspector only and does not certify structural adequacy or code compliance. The IFC file is the source artifact for downstream editing and any formal verification.

What happens when a build run fails?

A failed run preserves the last successful result and records the error in the verdict. Successful artifacts and measurements are stored in out/runs/<id>/, and out/latest.json names the current result.