IDS

You can run an AI-assisted MCP server that deterministically creates, validates, and manages IDS 1.0 files, ensuring full compliance while integrating with AI agents and IfcTester workflows. This server provides fast, session-based interactions and robust validation to keep IDS documents valid throughout the creation process.

How to use

You will run the IDS MCP server locally and connect your MCP client to it. The server exposes a collection of tools to create and manage IDS documents, add specifications, apply facets and restrictions, and perform validation. Use the server to produce IDS XML that is 100% compliant with the IDS 1.0 schema, export validated files, and inspect document structure and metadata.

How to install

Prerequisites you need installed on your machine before starting:

• Python 3.9+ is required
• pip is available with Python installations

Install and run the server with the following steps.

# Clone the project repository
git clone https://github.com/Quasar-Consulting-Group/ifc-ids-mcp.git
cd ifc-ids-mcp

# Install dependencies
pip install -r requirements.txt

# Install in development mode
pip install -e .

Configured MCP server connection (example)

You can connect a client via a local stdio configuration to run the server using Python in module mode. The following example shows how to start the IDS MCP server from a client configuration.

{
  "mcpServers": {
    "ids_mcp": {
      "command": "python",
      "args": ["-m", "ids_mcp_server"],
      "env": {
        "IDS_LOG_LEVEL": "INFO"
      }
    }
  }
}

Usage pattern with the MCP client

Start the server as shown above, then use your MCP client to interact with the server through the MCP protocol. You will create IDS documents, add specifications and facets, run validations, and export IDS XML files that conform to IDS 1.0.

Development and testing notes

The server is designed with test-driven development and deterministic output. Run tests to verify behavior, ensure 95%+ coverage, and validate XSD compliance during export.

Notes on early validation

Early validation provides immediate feedback when constraints are violated, helping AI agents correct invalid states before export. You will receive actionable error messages that point to the exact constraint and how to fix it.

Available tools

create_ids

Create a new IDS document with initial metadata and structure.

load_ids

Load an existing IDS document from file or XML string into the MCP session.

export_ids

Export the IDS document to XML with 100% XSD validation.

get_ids_info

Retrieve document structure and metadata for inspection.

add_specification

Add a new specification with IFC version and cardinality constraints.

add_entity_facet

Add an IFC entity facet to a specification’s applicability (subject to IDS 1.0 constraints).

add_property_facet

Add a property facet with a specified property set for validation.

add_attribute_facet

Add an IFC attribute facet requirement.

add_classification_facet

Add a classification facet constraint.

add_material_facet

Add material facet constraints.

add_partof_facet

Add spatial relationship facets.

add_enumeration_restriction

Constrain facet values to a predefined enumeration.

add_pattern_restriction

Constrain facet values with a regular expression pattern.

add_bounds_restriction

Constrain numeric facets within a range.

add_length_restriction

Constrain string length for facets.

validate_ids

Validate the IDS document against the IDS 1.0 XSD schema.

validate_ifc_model

Validate the linked IFC model against IDS constraints.