cad-to-data

cad-to-data is a skill for Claude Code, Codex from datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction. It costs 29 tokens per session (4,415 once invoked), scanned A, original, MIT.

A converter for CAD and BIM files into structured project data. CAD files contain technical drawings, while BIM files contain digital building models with elements such as walls, doors, pipes, and equipment.

In plain words
What is it for?
Use it to extract building elements, properties, quantities, and relationships from Revit, IFC, DWG, DGN, and other supported design formats.
Why use it?
It makes information buried in design files easier to analyze and reuse. This can reduce manual extraction of element properties, quantities, and relationships.

Skill for Claude CodeCodex

Which agent this was written for is unclear — built for openclaw. Also seen: built for openclaw.

Good fit Use it to extract building elements, properties, quantities, and relationships from Revit, IFC, DWG, DGN, and other supported design formats.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/cad-to-data
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill cad-to-data
Clone the repo
git clone --depth 1 https://github.com/datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for cad-to-data

README.md
[![agentmods](https://agentmods.dev/badge/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/cad-to-data/github.svg)](https://agentmods.dev/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/cad-to-data)
Your own site
<a href="https://agentmods.dev/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/cad-to-data"><img src="https://agentmods.dev/badge/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/cad-to-data/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for cad-to-data

Your own site · 80×15
<a href="https://agentmods.dev/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/cad-to-data"><img src="https://agentmods.dev/badge/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/cad-to-data.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 29 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,415 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00029 $0.04415
Opus 5 $0.00015 $0.02207
Sonnet 5 $0.00006 $0.00883
Haiku 4.5 $0.00003 $0.00441

Measured 9d ago against content hash 2867882b04e8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

cad-to-data scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

2_DDC_Book/2.4-PDF-CAD-to-Data/cad-to-data/SKILL.md · 655 lines

How it starts

The opening of the file, as written. The whole thing — 655 lines — stays where its author put it; the contents beside it link to each section on GitHub.

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"

@dataclass
class Point3D:
    """3D point"""
    x: float
    y: float
    z: float

@dataclass
class BoundingBox3D:
    """3D bounding box"""
    min_point: Point3D
    max_point: Point3D

    @property
    def width(self) -> float:
        return abs(self.max_point.x - self.min_point.x)

    @property
    def depth(self) -> float:
        return abs(self.max_point.y - self.min_point.y)

    @property
    def height(self) -> float:
        return abs(self.max_point.z - self.min_point.z)

    @property
    def volume(self) -> float:
        return self.width * self.depth * self.height

@dataclass
class MaterialInfo:
    """Material information"""
    name: str
    category: str
    color: Optional[str] = None
    area: float = 0.0
    volume: float = 0.0
    properties: Dict[str, Any] = field(default_factory=dict)

@dataclass
class CADElement:
    """Extracted CAD/BIM element"""
    id: str
    guid: str
    name: str
    category: ElementCategory
    type_name: str
    level: Optional[str] = None
    bounding_box: Optional[BoundingBox3D] = None
    properties: Dict[str, Any] = field(default_factory=dict)
    quantities: Dict[str, float] = field(default_factory=dict)
    materials: List[MaterialInfo] = field(default_factory=list)
    relationships: Dict[str, List[str]] = field(default_factory=dict)

@dataclass
class CADLayer:
    """CAD layer information"""
    name: str
    color: Optional[str] = None
    line_type: Optional[str] = None
    visible: bool = True
    element_count: int = 0

@dataclass
class CADExtractionResult:
    """Result of CAD extraction"""
    file_path: str
    file_format: CADFormat
    elements: List[CADElement]
    layers: List[CADLayer]
    levels: List[str]
    total_elements: int
    categories: Dict[str, int]
    extraction_time: float
    metadata: Dict[str, Any] = field(default_factory=dict)


class IFCExtractor:
    """Extract data from IFC files"""

    def __init__(self):
        self.schema_version = "IFC4"
        self.element_mapping = self._build_element_mapping()

    def _build_element_mapping(self) -> Dict[str, ElementCategory]:
        """Map IFC types to categories"""
        return {
            "IfcWall": ElementCategory.WALL,
            "IfcWallStandardCase": ElementCategory.WALL,
            "IfcSlab": ElementCategory.FLOOR,
            "IfcRoof": ElementCategory.ROOF,
            "IfcCeiling": ElementCategory.CEILING,
            "IfcDoor": ElementCategory.DOOR,
            "IfcWindow": ElementCategory.WINDOW,
            "IfcColumn": ElementCategory.COLUMN,
            "IfcBeam": ElementCategory.BEAM,
            "IfcStair": ElementCategory.STAIR,
            "IfcRamp": ElementCategory.RAMP,
            "IfcFurnishingElement": ElementCategory.FURNITURE,
            "IfcPipeSegment": ElementCategory.PIPE,
            "IfcDuctSegment": ElementCategory.DUCT,
            "IfcCableCarrierSegment": ElementCategory.CABLE_TRAY,
            "IfcSpace": ElementCategory.SPACE,
        }

    def extract(
        self,
        file_path: str,
        categories: Optional[List[ElementCategory]] = None
    ) -> CADExtractionResult:
        """
        Extract data from IFC file.

        Args:
            file_path: Path to IFC file
            categories: Optional filter for categories

        Returns:
            Extraction result
        """
        start_time = datetime.now()

        # In production, use ifcopenshell:
        # import ifcopenshell
        # ifc_file = ifcopenshell.open(file_path)

        # Simulated extraction
        elements = self._simulate_ifc_elements()

        # Filter by category if specified
        if categories:
            elements = [e for e in elements if e.category in categories]

        # Build category counts
        category_counts = {}
        for element in elements:
            cat = element.category.value
            category_counts[cat] = category_counts.get(cat, 0) + 1

        # Extract levels
        levels = list(set(e.level for e in elements if e.level))

        extraction_time = (datetime.now() - start_time).total_seconds()

        return CADExtractionResult(
            file_path=file_path,
            file_format=CADFormat.IFC,
            elements=elements,
            layers=[],  # IFC doesn't use layers in traditional sense
            levels=levels,
            total_elements=len(elements),
            categories=category_counts,
            extraction_time=extraction_time,
            metadata={
                "schema": self.schema_version,
                "project_name": "Sample Project"
            }
        )

    def _simulate_ifc_elements(self) -> List[CADElement]:
        """Simulate IFC element extraction"""
        elements = []

        # Sample walls
        for i in range(10):
            elements.append(CADElement(
                id=f"wall_{i}",
                guid=f"1234567890ABCDEF{i:04d}",
                name=f"Basic Wall {i}",
                category=ElementCategory.WALL,
                type_name="Basic Wall:200mm Concrete",
                level="Level 1",
                bounding_box=BoundingBox3D(
                    min_point=Point3D(i * 5, 0, 0),
                    max_point=Point3D(i * 5 + 5, 0.2, 3)
                ),
                properties={
                    "IsExternal": True,
                    "FireRating": "1 HR",
                    "LoadBearing": True
                },
                quantities={
                    "Length": 5.0,
                    "Height": 3.0,
                    "Width": 0.2,
                    "Area": 15.0,
                    "Volume": 3.0
                },
                materials=[
                    MaterialInfo(
                        name="Concrete",
                        category="Concrete",
                        area=15.0,
                        volume=3.0
                    )
                ]
            ))

        # Sample doors
        for i in range(5):
            elements.append(CADElement(
                id=f"door_{i}",
                guid=f"DOOR0000000000{i:04d}",
                name=f"Single Door {i}",
                category=ElementCategory.DOOR,
                type_name="Single Flush:900x2100",
                level="Level 1",
                properties={
                    "FireRating": "None",
                    "IsExternal": False
                },
                quantities={
                    "Width": 0.9,
                    "Height": 2.1,
                    "Area": 1.89
                },
                relationships={
                    "host_wall": [f"wall_{i}"]
                }
            ))

        # Sample spaces
        for i in range(3):
            elements.append(CADElement(
                id=f"space_{i}",
                guid=f"SPACE000000000{i:04d}",
                name=f"Room {i+101}",
                category=ElementCategory.SPACE,
                type_name="Office",
                level="Level 1",
                quantities={
                    "Area": 25.0 + i * 5,
                    "Volume": 75.0 + i * 15,
                    "Perimeter": 20.0 + i * 2
                },
                properties={
                    "OccupancyType": "Office",
                    "DesignOccupancy": 4
                }
            ))

        return elements

    def get_quantities(
        self,
        elements: List[CADElement],
        quantity_type: str = "all"
    ) -> Dict[str, float]:
        """Aggregate quantities from elements"""
        totals = {}

        for element in elements:
            for qty_name, qty_value in element.quantities.items():
                if quantity_type == "all" or qty_name.lower() == quantity_type.lower():
                    key = f"{element.category.value}_{qty_name}"
                    totals[key] = totals.get(key, 0) + qty_value

        return totals


class DWGExtractor:
    """Extract data from DWG/DXF files"""

    def __init__(self):
        self.supported_entities = ["LINE", "POLYLINE", "CIRCLE", "ARC", "TEXT", "MTEXT", "INSERT", "HATCH"]

    def extract(
        self,
        file_path: str,
        layers: Optional[List[str]] = None
    ) -> CADExtractionResult:
        """Extract data from DWG file"""
        start_time = datetime.now()

        # In production, use ezdxf:
        # import ezdxf
        # doc = ezdxf.readfile(file_path)

        # Simulated extraction
        elements, cad_layers = self._simulate_dwg_extraction()

        # Filter by layers if specified
        if layers:
            elements = [e for e in elements if e.properties.get("layer") in layers]

        extraction_time = (datetime.now() - start_time).total_seconds()

        return CADExtractionResult(
            file_path=file_path,
            file_format=CADFormat.DWG,
            elements=elements,
            layers=cad_layers,
            levels=[],
            total_elements=len(elements),
            categories={"generic": len(elements)},
            extraction_time=extraction_time,
            metadata={"units": "millimeters"}
        )

    def _simulate_dwg_extraction(self) -> Tuple[List[CADElement], List[CADLayer]]:
        """Simulate DWG extraction"""
        elements = []
        layers = [
            CADLayer("Walls", "Red", "Continuous", True, 15),
            CADLayer("Doors", "Blue", "Continuous", True, 8),
            CADLayer("Windows", "Cyan", "Continuous", True, 12),
            CADLayer("Dimensions", "Green", "Continuous", True, 50),
            CADLayer("Text", "White", "Continuous", True, 25),
        ]

        # Simulate polylines (walls)
        for i in range(15):
            elements.append(CADElement(
                id=f"polyline_{i}",
                guid=f"PL{i:08d}",
                name=f"Polyline {i}",
                category=ElementCategory.GENERIC,
                type_name="POLYLINE",
                properties={
                    "layer": "Walls",
                    "color": "Red",
                    "closed": True
                },
                quantities={
                    "Length": 10.5 + i * 0.5
                }
            ))

        return elements, layers


class CADDataConverter:
    """
    Convert CAD/BIM files to structured data.
    Based on DDC methodology Chapter 2.4.
    """

    def __init__(self):
        self.ifc_extractor = IFCExtractor()
        self.dwg_extractor = DWGExtractor()

    def convert(
        self,
        file_path: str,
        output_format: str = "json"
    ) -> Dict[str, Any]:
        """
        Convert CAD file to structured data.

        Args:
            file_path: Path to CAD file
            output_format: Output format (json, csv, dataframe)

        Returns:
            Structured data
        """
        # Detect file format
        file_format = self._detect_format(file_path)

        # Extract based on format
        if file_format == CADFormat.IFC:
            result = self.ifc_extractor.extract(file_path)
        elif file_format in [CADFormat.DWG, CADFormat.DXF]:
            result = self.dwg_extractor.extract(file_path)
        else:
            raise ValueError(f"Unsupported format: {file_format}")

        # Convert to output format
        return self._format_output(result, output_format)

    def _detect_format(self, file_path: str) -> CADFormat:
        """Detect CAD file format"""
        extension = file_path.lower().split(".")[-1]

        format_map = {
            "ifc": CADFormat.IFC,
            "rvt": CADFormat.RVT,
            "dwg": CADFormat.DWG,
            "dxf": CADFormat.DXF,
            "dgn": CADFormat.DGN,
            "nwd": CADFormat.NWD,
        }

        return format_map.get(extension, CADFormat.IFC)

    def _format_output(
        self,
        result: CADExtractionResult,
        format: str
    ) -> Dict[str, Any]:
        """Format extraction result"""
        output = {
            "file": result.file_path,
            "format": result.file_format.value,
            "total_elements": result.total_elements,
            "categories": result.categories,
            "levels": result.levels,
            "extraction_time": result.extraction_time,
            "elements": []
        }

        for element in result.elements:
            output["elements"].append({
                "id": element.id,
                "guid": element.guid,
                "name": element.name,
                "category": element.category.value,
                "type": element.type_name,
                "level": element.level,
                "properties": element.properties,
                "quantities": element.quantities,
                "materials": [
                    {"name": m.name, "area": m.area, "volume": m.volume}
                    for m in element.materials
                ]
            })

        return output

    def extract_quantities(
        self,
        file_path: str,
        categories: Optional[List[ElementCategory]] = None
    ) -> Dict[str, Any]:
        """Extract quantity takeoff from CAD file"""
        file_format = self._detect_format(file_path)

        if file_format == CADFormat.IFC:
            result = self.ifc_extractor.extract(file_path, categories)
        else:
            result = self.dwg_extractor.extract(file_path)

        # Aggregate quantities by category
        quantities = {}
        for element in result.elements:
            cat = element.category.value
            if cat not in quantities:
                quantities[cat] = {
                    "count": 0,
                    "totals": {}
                }

            quantities[cat]["count"] += 1

            for qty_name, qty_value in element.quantities.items():
                if qty_name not in quantities[cat]["totals"]:
                    quantities[cat]["totals"][qty_name] = 0
                quantities[cat]["totals"][qty_name] += qty_value

        return {
            "file": file_path,
            "quantities": quantities,
            "summary": {
                "total_elements": result.total_elements,
                "categories": list(quantities.keys())
            }
        }

    def extract_schedule(
        self,
        file_path: str,
        category: ElementCategory,
        fields: List[str]
    ) -> List[Dict]:
        """Extract schedule data for specific category"""
        file_format = self._detect_format(file_path)

        if file_format == CADFormat.IFC:
            result = self.ifc_extractor.extract(file_path, [category])
        else:
            result = self.dwg_extractor.extract(file_path)

        schedule = []
        for element in result.elements:
            if element.category == category:
                row = {"id": element.id, "name": element.name, "type": element.type_name}

                for field in fields:
                    if field in element.properties:
                        row[field] = element.properties[field]
                    elif field in element.quantities:
                        row[field] = element.quantities[field]

                schedule.append(row)

        return schedule

    def export_to_json(
        self,
        result: CADExtractionResult,
        output_path: str
    ):
        """Export extraction result to JSON file"""
        output = self._format_output(result, "json")

        with open(output_path, 'w') as f:
            json.dump(output, f, indent=2)

    def generate_report(self, result: CADExtractionResult) -> str:
        """Generate extraction report"""
        report = f"""
# CAD Extraction Report

**File:** {result.file_path}
**Format:** {result.file_format.value}
**Total Elements:** {result.total_elements}
**Extraction Time:** {result.extraction_time:.2f}s

## Elements by Category
"""
        for cat, count in result.categories.items():
            report += f"- **{cat.title()}:** {count}\n"

        if result.levels:
            report += "\n## Levels\n"
            for level in result.levels:
                report += f"- {level}\n"

        if result.layers:
            report += "\n## Layers\n"
            for layer in result.layers:
                report += f"- {layer.name}: {layer.element_count} elements\n"

        return report

Read the full file on GitHub · 655 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 655 lines · 29 tokens per session scan A 2867882b04e8

Subscribe to this mod's changes

cad-to-data is a skill published in the GitHub repository datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction (310 stars, last pushed 21d ago), licensed MIT. It adds 29 tokens to every session and 4,415 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other skills, from other repositories

tflite-micro-integration

Use when integrating or debugging TensorFlow Lite Micro (LiteRT) on MCUs — op resolver, tensor arena sizing, AllocateTensors, Invoke, int8 quantization, and missing-op errors.

easyzoom/aix-skills · 47 tokens

tinymaix-integration

Use when integrating, porting, configuring, or debugging TinyMaix MCU inference, model loading, tensor shapes, memory, quantization, or TinyML runtime issues.

easyzoom/aix-skills · 39 tokens

mk:llms

Generate llms.txt files from project documentation following the llmstxt.org spec. Use when asked to create AI-friendly documentation indexes, generate llms.txt, or make a project discoverable by AI assistants.

ngocsangyem/MeowKit · 46 tokens

disaster-prediction

Analyze disaster prediction and early warning systems — model accuracy for flood, earthquake, wildfire, hurricane, and tsunami hazards, data pipeline reliability from sensor networks and satellite feeds, alert distribution latency and channel coverage.

tinh2/skills-hub-registry · 46 tokens

drug-discovery-ops

Audit a drug discovery pipeline for operational efficiency and scientific rigor. Triggers: building or reviewing pharma R&D platforms, cheminformatics pipelines, or screening data management systems.

tinh2/skills-hub-registry · 40 tokens

experiment-tracking

Audit ML experiment tracking infrastructure for reproducibility gaps, parameter logging completeness, metric capture, artifact management, and pipeline orchestration. Covers MLflow, Weights and Biases, DVC, Sacred, Neptune, Hydra configs, model registries, and produces a reproducibility scorecard (0-30) with…

tinh2/skills-hub-registry · 73 tokens