bim-clash-detection

bim-clash-detection is a skill for Claude Code, Codex from datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction. It costs 28 tokens per session (3,114 once invoked), scanned A, original, MIT.

A system for finding geometric conflicts in BIM models, which are digital 3D models of buildings and their systems. It checks architectural, structural, and building-services elements before construction.

In plain words
What is it for?
Use it to detect, categorize, prioritize, and analyze clashes between systems such as mechanical, electrical, plumbing, structural, and architectural components.
Why use it?
It helps find physical intersections, clearance problems, duplicated elements, and sequencing conflicts before they cause site rework or late design changes.

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 detect, categorize, prioritize, and analyze clashes between systems such as mechanical, electrical, plumbing, structural, and architectural components.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/bim-clash-detection
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 bim-clash-detection
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 bim-clash-detection

README.md
[![agentmods](https://agentmods.dev/badge/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/bim-clash-detection/github.svg)](https://agentmods.dev/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/bim-clash-detection)
Your own site
<a href="https://agentmods.dev/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/bim-clash-detection"><img src="https://agentmods.dev/badge/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/bim-clash-detection/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 bim-clash-detection

Your own site · 80×15
<a href="https://agentmods.dev/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/bim-clash-detection"><img src="https://agentmods.dev/badge/skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/bim-clash-detection.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 28 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,114 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 $0.00028 $0.03114
Opus 5 $0.00014 $0.01557
Sonnet 5 $0.00006 $0.00623
Haiku 4.5 $0.00003 $0.00311

Measured 12d ago against content hash 5abe027f0db2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

bim-clash-detection 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 12d 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:

1_DDC_Toolkit/BIM-Analysis/bim-clash-detection/SKILL.md · 401 lines

How it starts

The opening of the file, as written. The whole thing — 401 lines — stays where its author put it; the contents beside it link to each section on GitHub.

BIM Clash Detection

Business Case

Problem Statement

Coordination issues cause significant rework:

  • MEP vs structural conflicts discovered on site
  • Late design changes increase costs
  • Manual clash review is time-consuming
  • No standardized clash categorization

Solution

Automated clash detection and analysis system that identifies conflicts between building systems and provides prioritized resolution recommendations.

Business Value

  • Cost savings - Detect issues before construction
  • Time reduction - Automated clash identification
  • Better coordination - Systematic conflict resolution
  • Quality improvement - Fewer field issues

Technical Implementation

import pandas as pd
from datetime import datetime
from typing import Dict, Any, List, Optional, Tuple
from dataclasses import dataclass, field
from enum import Enum
import math


class ClashType(Enum):
    """Types of clashes."""
    HARD = "hard"           # Physical intersection
    SOFT = "soft"           # Clearance violation
    WORKFLOW = "workflow"   # Sequencing conflict
    DUPLICATE = "duplicate" # Duplicated elements


class ClashStatus(Enum):
    """Clash resolution status."""
    NEW = "new"
    ACTIVE = "active"
    RESOLVED = "resolved"
    APPROVED = "approved"
    IGNORED = "ignored"


class ClashSeverity(Enum):
    """Clash severity level."""
    CRITICAL = "critical"
    MAJOR = "major"
    MINOR = "minor"
    INFO = "info"


class Discipline(Enum):
    """BIM disciplines."""
    ARCHITECTURAL = "architectural"
    STRUCTURAL = "structural"
    MECHANICAL = "mechanical"
    ELECTRICAL = "electrical"
    PLUMBING = "plumbing"
    FIRE_PROTECTION = "fire_protection"
    CIVIL = "civil"


@dataclass
class BoundingBox:
    """3D bounding box."""
    min_x: float
    min_y: float
    min_z: float
    max_x: float
    max_y: float
    max_z: float

    def intersects(self, other: 'BoundingBox') -> bool:
        """Check if boxes intersect."""
        return (self.min_x <= other.max_x and self.max_x >= other.min_x and
                self.min_y <= other.max_y and self.max_y >= other.min_y and
                self.min_z <= other.max_z and self.max_z >= other.min_z)

    def volume(self) -> float:
        """Calculate bounding box volume."""
        return ((self.max_x - self.min_x) *
                (self.max_y - self.min_y) *
                (self.max_z - self.min_z))

    def center(self) -> Tuple[float, float, float]:
        """Get center point."""
        return (
            (self.min_x + self.max_x) / 2,
            (self.min_y + self.max_y) / 2,
            (self.min_z + self.max_z) / 2
        )


@dataclass
class BIMElement:
    """BIM element representation."""
    element_id: str
    name: str
    discipline: Discipline
    category: str  # e.g., "Duct", "Beam", "Pipe"
    level: str
    bounding_box: BoundingBox
    properties: Dict[str, Any] = field(default_factory=dict)

    def distance_to(self, other: 'BIMElement') -> float:
        """Calculate distance between element centers."""
        c1 = self.bounding_box.center()
        c2 = other.bounding_box.center()
        return math.sqrt(
            (c2[0] - c1[0])**2 +
            (c2[1] - c1[1])**2 +
            (c2[2] - c1[2])**2
        )


@dataclass
class Clash:
    """Clash between two elements."""
    clash_id: str
    element_a: BIMElement
    element_b: BIMElement
    clash_type: ClashType
    severity: ClashSeverity
    status: ClashStatus
    distance: float  # Penetration depth (negative) or clearance gap
    location: Tuple[float, float, float]
    detected_at: datetime
    resolved_at: Optional[datetime] = None
    assigned_to: Optional[str] = None
    notes: str = ""

    def to_dict(self) -> Dict[str, Any]:
        return {
            'clash_id': self.clash_id,
            'element_a_id': self.element_a.element_id,
            'element_a_name': self.element_a.name,
            'element_a_discipline': self.element_a.discipline.value,
            'element_b_id': self.element_b.element_id,
            'element_b_name': self.element_b.name,
            'element_b_discipline': self.element_b.discipline.value,
            'clash_type': self.clash_type.value,
            'severity': self.severity.value,
            'status': self.status.value,
            'distance': round(self.distance, 3),
            'location_x': self.location[0],
            'location_y': self.location[1],
            'location_z': self.location[2],
            'level': self.element_a.level,
            'detected_at': self.detected_at.isoformat(),
            'assigned_to': self.assigned_to,
            'notes': self.notes
        }


@dataclass
class ClashTest:
    """Clash test configuration."""
    name: str
    discipline_a: Discipline
    discipline_b: Discipline
    clash_type: ClashType
    tolerance: float = 0.0  # Clearance tolerance in meters
    enabled: bool = True


class BIMClashDetector:
    """Detect and manage BIM clashes."""

    def __init__(self):
        self.elements: List[BIMElement] = []
        self.clashes: List[Clash] = []
        self.clash_tests: List[ClashTest] = []
        self._clash_counter = 0

    def load_elements(self, elements_df: pd.DataFrame) -> int:
        """Load BIM elements from DataFrame."""
        loaded = 0
        for _, row in elements_df.iterrows():
            element = BIMElement(
                element_id=str(row.get('element_id', '')),
                name=str(row.get('name', '')),
                discipline=Discipline(row.get('discipline', 'architectural')),
                category=str(row.get('category', '')),
                level=str(row.get('level', '')),
                bounding_box=BoundingBox(
                    min_x=float(row.get('min_x', 0)),
                    min_y=float(row.get('min_y', 0)),
                    min_z=float(row.get('min_z', 0)),
                    max_x=float(row.get('max_x', 0)),
                    max_y=float(row.get('max_y', 0)),
                    max_z=float(row.get('max_z', 0))
                )
            )
            self.elements.append(element)
            loaded += 1
        return loaded

    def add_clash_test(self, test: ClashTest):
        """Add clash test configuration."""
        self.clash_tests.append(test)

    def setup_standard_tests(self):
        """Setup standard MEP coordination tests."""
        standard_tests = [
            ClashTest("MEP vs Structure", Discipline.MECHANICAL, Discipline.STRUCTURAL, ClashType.HARD),
            ClashTest("Electrical vs Structure", Discipline.ELECTRICAL, Discipline.STRUCTURAL, ClashType.HARD),
            ClashTest("Plumbing vs Structure", Discipline.PLUMBING, Discipline.STRUCTURAL, ClashType.HARD),
            ClashTest("MEP vs MEP", Discipline.MECHANICAL, Discipline.ELECTRICAL, ClashType.HARD),
            ClashTest("Duct Clearance", Discipline.MECHANICAL, Discipline.MECHANICAL, ClashType.SOFT, tolerance=0.05),
            ClashTest("Fire Protection", Discipline.FIRE_PROTECTION, Discipline.STRUCTURAL, ClashType.HARD),
        ]
        for test in standard_tests:
            self.add_clash_test(test)

    def run_clash_detection(self) -> List[Clash]:
        """Run all clash tests."""
        new_clashes = []

        for test in self.clash_tests:
            if not test.enabled:
                continue

            # Filter elements by discipline
            elements_a = [e for e in self.elements if e.discipline == test.discipline_a]
            elements_b = [e for e in self.elements if e.discipline == test.discipline_b]

            # Check all pairs
            for elem_a in elements_a:
                for elem_b in elements_b:
                    if elem_a.element_id == elem_b.element_id:
                        continue

                    clash = self._check_clash(elem_a, elem_b, test)
                    if clash:
                        new_clashes.append(clash)

        self.clashes.extend(new_clashes)
        return new_clashes

    def _check_clash(self, elem_a: BIMElement, elem_b: BIMElement,
                     test: ClashTest) -> Optional[Clash]:
        """Check if two elements clash."""

        # Expand bounding box by tolerance for soft clashes
        box_a = elem_a.bounding_box
        box_b = elem_b.bounding_box

        if test.clash_type == ClashType.SOFT:
            # Add clearance tolerance
            expanded_a = BoundingBox(
                box_a.min_x - test.tolerance, box_a.min_y - test.tolerance, box_a.min_z - test.tolerance,
                box_a.max_x + test.tolerance, box_a.max_y + test.tolerance, box_a.max_z + test.tolerance
            )
            intersects = expanded_a.intersects(box_b)
        else:
            intersects = box_a.intersects(box_b)

        if not intersects:
            return None

        # Calculate clash point and severity
        self._clash_counter += 1
        clash_id = f"CLH-{self._clash_counter:05d}"

        # Clash location (center of intersection)
        location = (
            (max(box_a.min_x, box_b.min_x) + min(box_a.max_x, box_b.max_x)) / 2,
            (max(box_a.min_y, box_b.min_y) + min(box_a.max_y, box_b.max_y)) / 2,
            (max(box_a.min_z, box_b.min_z) + min(box_a.max_z, box_b.max_z)) / 2
        )

        # Calculate penetration depth
        distance = elem_a.distance_to(elem_b)

        # Determine severity
        if test.clash_type == ClashType.HARD:
            severity = ClashSeverity.CRITICAL if distance < 0.1 else ClashSeverity.MAJOR
        else:
            severity = ClashSeverity.MINOR if distance > test.tolerance else ClashSeverity.MAJOR

        return Clash(
            clash_id=clash_id,
            element_a=elem_a,
            element_b=elem_b,
            clash_type=test.clash_type,
            severity=severity,
            status=ClashStatus.NEW,
            distance=distance,
            location=location,
            detected_at=datetime.now()
        )

    def get_summary(self) -> Dict[str, Any]:
        """Get clash detection summary."""
        by_severity = {}
        by_discipline = {}
        by_status = {}

        for clash in self.clashes:
            # By severity
            sev = clash.severity.value
            by_severity[sev] = by_severity.get(sev, 0) + 1

            # By discipline pair
            pair = f"{clash.element_a.discipline.value} vs {clash.element_b.discipline.value}"
            by_discipline[pair] = by_discipline.get(pair, 0) + 1

            # By status
            stat = clash.status.value
            by_status[stat] = by_status.get(stat, 0) + 1

        return {
            'total_clashes': len(self.clashes),
            'by_severity': by_severity,
            'by_discipline': by_discipline,
            'by_status': by_status,
            'elements_checked': len(self.elements),
            'tests_run': len([t for t in self.clash_tests if t.enabled])
        }

    def export_to_dataframe(self) -> pd.DataFrame:
        """Export clashes to DataFrame."""
        return pd.DataFrame([c.to_dict() for c in self.clashes])

    def resolve_clash(self, clash_id: str, resolution_note: str):
        """Mark clash as resolved."""
        for clash in self.clashes:
            if clash.clash_id == clash_id:
                clash.status = ClashStatus.RESOLVED
                clash.resolved_at = datetime.now()
                clash.notes = resolution_note
                break

    def assign_clash(self, clash_id: str, assignee: str):
        """Assign clash to team member."""
        for clash in self.clashes:
            if clash.clash_id == clash_id:
                clash.assigned_to = assignee
                clash.status = ClashStatus.ACTIVE
                break

Read the full file on GitHub · 401 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. 12d ago First seen · 401 lines · 28 tokens per session scan A 5abe027f0db2

Subscribe to this mod's changes

bim-clash-detection is a skill published in the GitHub repository datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction (310 stars, last pushed 20d ago), licensed MIT. It adds 28 tokens to every session and 3,114 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 2026-08-30.