Pulls structured data out of CAD and BIM files like Revit, IFC, DWG, and DGN formats. You get element properties, quantities, bounding boxes, material info, and relationships extracted into Python dataclasses. The code shows a clean type system with enums for categories (walls, doors, pipes, ducts) and proper 3D geometry classes. Built around something called DDC methodology from a book on data transformation. The IFC extractor is the main example here, with mapping logic for converting IFC types to standardized categories. Useful when you need to process construction models programmatically instead of clicking around in desktop software. The sample code is mostly simulation, so you'll need ifcopenshell or similar for real file parsing.

skills.sh installs
349
GitHub stars
401
last updated
Aug 21, 2026

Install

$ npx -y skills add datadrivenconstruction/ddc_skills_for_ai_agents_in_construction --skill cad-to-data --agent claude-code

Installs into .claude/skills of the current project.

Files

CAD To Data

Overview

Based on DDC methodology (Chapter 2.4), this skill converts CAD and BIM files to structured data, extracting element properties, quantities, and relationships from Revit, IFC, DWG, and DGN files.

Book Reference: "Преобразование данных в структурированную форму" / "Data Transformation to Structured Form"

Quick Start

from dataclasses import dataclass, field
from enum import Enum
from typing import List, Dict, Optional, Any, Tuple, Generator
from datetime import datetime
import json

class CADFormat(Enum):
    """Supported CAD/BIM formats"""
    IFC = "ifc"
    RVT = "rvt"
    DWG = "dwg"
    DXF = "dxf"
    DGN = "dgn"
    NWD = "nwd"
    STEP = "step"

class ElementCategory(Enum):
    """BIM element categories"""
    WALL = "wall"
    FLOOR = "floor"
    ROOF = "roof"
    CEILING = "ceiling"
    DOOR = "door"
    WINDOW = "window"
    COLUMN = "column"
    BEAM = "beam"
    STAIR = "stair"
    RAMP = "ramp"
    FURNITURE = "furniture"
    EQUIPMENT = "equipment"
    PIPE = "pipe"
    DUCT = "duct"
    CABLE_TRAY = "cable_tray"
    SPACE = "space"
    GENERIC = "generic"

Sponsors

Details

Repository
github.com/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction
Category
AI & Agent Building
First seen
Jun 3, 2026

More from datadrivenconstruction/ddc_skills_for_ai_agents_in_construction

All 3
1jwynia avatarconfig-scanjwynia/agent-skillsDetect security misconfigurations in config files, Docker, and IaC. Use when reviewing configuration security for containers, Kubernetes, Terraform, or applicat349skills.sh installs
2krutikjain avatarandroid-kotlin-corekrutikjain/android-agent-skillsUse Kotlin idioms safely in Android apps, including nullability, data classes, sealed types, extension functions, and collection pipelines.349skills.sh installs
3siviter-xyz avatarbackend-engineersiviter-xyz/dot-agentBuild robust backend systems with modern technologies (Node.js, Python, Go, Rust), frameworks (NestJS, FastAPI, Django), databases (PostgreSQL, MongoDB, Redis),349skills.sh installs
4specstoryai avatarspecstory-session-summaryspecstoryai/agent-skillsSummarize recent SpecStory AI coding sessions in standup format. Use when the user wants to review sessions from .specstory/history, prepare for standups, track349skills.sh installs
5automattic avatarwp-project-triageautomattic/agent-skillswp project triage347skills.sh installs
6gohypergiant avataraccelint-ts-performancegohypergiant/agent-skillsSystematic JavaScript/TypeScript performance audit and optimization using V8 profiling and runtime patterns. Use when (1) Users say 'optimize performance', 'aud347skills.sh installs