ifcMCP

IfcMCP is an MCP server that lets LLM agents talk to IFC (Industry Foundation Classes) files. It exposes a simple HTTP MCP interface so your AI agents can query IFC data through a stable, standardized protocol and receive structured results.

How to use

You will run the MCP server locally and connect your MCP client to it. Start by ensuring the server is running at http://127.0.0.1:8000/mcp. Your client can send requests to this endpoint to query IFC entities, properties, locations, and spatial relations. Use the supported tools to fetch entities, properties, openings, and space boundaries as part of your LLM workflows. Treat IFC data as a set of structured responses you can reason about, filter, and combine with other data sources.

How to install

Prerequisites you need before installation: Python installed on your system.

Step 1: Install required Python packages.

Step 2: Start the server from its directory.

Step 3: Connect your MCP client to the server using the provided URL.

Configuration and usage notes

Server URL for MCP clients: http://127.0.0.1:8000/mcp

Supported tools you can invoke through the MCP interface: get_entities, get_named_property_of_entities, get_entity_properties, get_entity_location, get_entities_in_spatial, get_openings_on_wall, get_space_boundaries

Available tools

get_entities

Fetch a list of IFC entities matching given criteria or filters.

get_named_property_of_entities

Retrieve named properties for a set of entities to extract meaningful attributes.

get_entity_properties

Return a full property set for a specified IFC entity including basic metadata.

get_entity_location

Obtain the spatial location or placement information for an IFC entity.

get_entities_in_spatial

Query entities within a defined spatial region or bounding geometry.

get_openings_on_wall

Identify openings on a wall and related openings data.

get_space_boundaries

Get boundaries and extents for a given space within the IFC model.