Automated safety compliance verification for construction sites. Check PPE usage, zone access, working at heights regulations, and generate compliance reports using rule-based and ML approaches.
npx skills add https://github.com/datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill safety-compliance-checker
This skill implements automated safety compliance checking for construction projects. Verify regulatory requirements, track safety metrics, and identify potential violations before they become incidents.
Compliance Areas:
from dataclasses import dataclass
from typing import List, Dict, Optional
from enum import Enum
from datetime import datetime, date
class ComplianceStatus(Enum):
COMPLIANT = "compliant"
NON_COMPLIANT = "non_compliant"
PARTIAL = "partial"
NOT_APPLICABLE = "not_applicable"
PENDING_REVIEW = "pending_review"
@dataclass
class ComplianceCheck:
rule_id: str
rule_name: str
status: ComplianceStatus
findings: List[str]
evidence: Optional[str]
checked_at: datetime
checked_by: str
# Quick compliance check
def check_ppe_compliance(workers: List[Dict]) -> List[ComplianceCheck]:
"""Check PPE compliance for workers"""
checks = []
for worker in workers:
findings = []
required_ppe = worker.get('required_ppe', ['helmet', 'vest', 'boots'])
actual_ppe = worker.get('actual_ppe', [])
missing = set(required_ppe) - set(actual_ppe)
if missing:
findings.append(f"Missing PPE: {', '.join(missing)}")
status = ComplianceStatus.COMPLIANT if not missing else ComplianceStatus.NON_COMPLIANT
checks.append(ComplianceCheck(
rule_id="PPE-001",
rule_name="Personal Protective Equipment",
status=status,
findings=findings,
evidence=f"Worker ID: {worker.get('id')}",
checked_at=datetime.now(),
checked_by="automated_system"
))
return checks
# Example usage
workers = [
{'id': 'W001', 'required_ppe': ['helmet', 'vest', 'boots'], 'actual_ppe': ['helmet', 'vest']},
{'id': 'W002', 'required_ppe': ['helmet', 'vest', 'boots'], 'actual_ppe': ['helmet', 'vest', 'boots']}
]
results = check_ppe_compliance(workers)
for r in results:
print(f"{r.evidence}: {r.status.value} - {r.findings}")
from dataclasses import dataclass, field
from typing import List, Dict, Callable, Optional, Any
from enum import Enum
from datetime import datetime, date, timedelta
import json
class RiskLevel(Enum):
LOW = 1
MEDIUM = 2
HIGH = 3
CRITICAL = 4
class RuleCategory(Enum):
PPE = "Personal Protective Equipment"
FALL_PROTECTION = "Fall Protection"
ELECTRICAL = "Electrical Safety"
EXCAVATION = "Excavation Safety"
CONFINED_SPACE = "Confined Space"
HOT_WORK = "Hot Work"
FIRE = "Fire Prevention"
HAZMAT = "Hazardous Materials"
CRANE = "Crane & Lifting"
SCAFFOLDING = "Scaffolding"
@dataclass
class SafetyRule:
rule_id: str
name: str
description: str
category: RuleCategory
risk_level: RiskLevel
regulation_ref: str # OSHA, local codes
check_function: Optional[Callable] = None
parameters: Dict = field(default_factory=dict)
@dataclass
class Violation:
rule: SafetyRule
location: str
description: str
detected_at: datetime
severity: RiskLevel
corrective_action: str
deadline: date
status: str = "open"
assigned_to: Optional[str] = None
class SafetyComplianceEngine:
"""Rule-based safety compliance checking engine"""
def __init__(self):
self.rules: Dict[str, SafetyRule] = {}
self.violations: List[Violation] = []
self._load_default_rules()
def _load_default_rules(self):
"""Load default safety rules"""
default_rules = [
SafetyRule(
rule_id="OSHA-1926.100",
name="Head Protection",
description="Employees working in areas where there is a possible danger of head injury shall wear protective helmets",
category=RuleCategory.PPE,
risk_level=RiskLevel.HIGH,
regulation_ref="OSHA 29 CFR 1926.100",
parameters={"required_ppe": ["helmet"]}
),
SafetyRule(
rule_id="OSHA-1926.501",
name="Fall Protection - General",
description="Each employee on walking/working surfaces with unprotected sides 6 feet or more above lower level shall be protected",
category=RuleCategory.FALL_PROTECTION,
risk_level=RiskLevel.CRITICAL,
regulation_ref="OSHA 29 CFR 1926.501",
parameters={"height_threshold_ft": 6}
),
SafetyRule(
rule_id="OSHA-1926.651",
name="Excavation General Requirements",
description="Daily inspections of excavations, adjacent areas, and protective systems",
category=RuleCategory.EXCAVATION,
risk_level=RiskLevel.HIGH,
regulation_ref="OSHA 29 CFR 1926.651",
parameters={"inspection_frequency": "daily"}
),
SafetyRule(
rule_id="OSHA-1926.1200",
name="Confined Space Entry",
description="Permit-required confined space program for construction",
category=RuleCategory.CONFINED_SPACE,
risk_level=RiskLevel.CRITICAL,
regulation_ref="OSHA 29 CFR 1926.1200",
parameters={"permit_required": True}
),
SafetyRule(
rule_id="OSHA-1926.352",
name="Fire Prevention - Welding",
description="Fire watch and extinguisher required for hot work",
category=RuleCategory.HOT_WORK,
risk_level=RiskLevel.HIGH,
regulation_ref="OSHA 29 CFR 1926.352",
parameters={"fire_watch_duration_min": 30}
),
SafetyRule(
rule_id="OSHA-1926.451",
name="Scaffolding General Requirements",
description="Scaffold platforms shall be fully planked between front uprights and guardrail supports",
category=RuleCategory.SCAFFOLDING,
risk_level=RiskLevel.HIGH,
regulation_ref="OSHA 29 CFR 1926.451",
parameters={"max_height_without_tie": 26}
),
SafetyRule(
rule_id="OSHA-1926.405",
name="Electrical Wiring Methods",
description="All electrical equipment and circuits shall be grounded",
category=RuleCategory.ELECTRICAL,
risk_level=RiskLevel.CRITICAL,
regulation_ref="OSHA 29 CFR 1926.405",
parameters={"gfci_required": True}
)
]
for rule in default_rules:
self.rules[rule.rule_id] = rule
def add_rule(self, rule: SafetyRule):
"""Add custom safety rule"""
self.rules[rule.rule_id] = rule
def check_work_activity(self, activity: Dict) -> List[ComplianceCheck]:
"""Check compliance for a work activity"""
checks = []
activity_type = activity.get('type', '')
location = activity.get('location', '')
height = activity.get('height_ft', 0)
workers = activity.get('workers', [])
permits = activity.get('permits', [])
equipment = activity.get('equipment', [])
# Fall protection check
if height >= 6:
fall_rule = self.rules.get("OSHA-1926.501")
if fall_rule:
has_protection = any(
'harness' in w.get('ppe', []) or 'guardrail' in equipment
for w in workers
)
checks.append(ComplianceCheck(
rule_id=fall_rule.rule_id,
rule_name=fall_rule.name,
status=ComplianceStatus.COMPLIANT if has_protection else ComplianceStatus.NON_COMPLIANT,
findings=[] if has_protection else [f"Working at {height}ft without fall protection"],
evidence=f"Activity: {activity_type} at {location}",
checked_at=datetime.now(),
checked_by="automated_system"
))
# PPE check for all workers
for worker in workers:
ppe_rule = self.rules.get("OSHA-1926.100")
if ppe_rule:
has_helmet = 'helmet' in worker.get('ppe', [])
checks.append(ComplianceCheck(
rule_id=ppe_rule.rule_id,
rule_name=ppe_rule.name,
status=ComplianceStatus.COMPLIANT if has_helmet else ComplianceStatus.NON_COMPLIANT,
findings=[] if has_helmet else ["Missing hard hat"],
evidence=f"Worker: {worker.get('id')}",
checked_at=datetime.now(),
checked_by="automated_system"
))
# Hot work permit check
if activity_type in ['welding', 'cutting', 'brazing']:
hot_work_rule = self.rules.get("OSHA-1926.352")
if hot_work_rule:
has_permit = 'hot_work' in permits
has_fire_watch = activity.get('fire_watch', False)
has_extinguisher = 'fire_extinguisher' in equipment
findings = []
if not has_permit:
findings.append("Missing hot work permit")
if not has_fire_watch:
findings.append("No fire watch assigned")
if not has_extinguisher:
findings.append("No fire extinguisher on site")
checks.append(ComplianceCheck(
rule_id=hot_work_rule.rule_id,
rule_name=hot_work_rule.name,
status=ComplianceStatus.COMPLIANT if not findings else ComplianceStatus.NON_COMPLIANT,
findings=findings,
evidence=f"Activity: {activity_type} at {location}",
checked_at=datetime.now(),
checked_by="automated_system"
))
# Confined space check
if activity.get('confined_space', False):
cs_rule = self.rules.get("OSHA-1926.1200")
if cs_rule:
has_permit = 'confined_space' in permits
has_attendant = activity.get('attendant', False)
atmospheric_tested = activity.get('atmospheric_test', False)
findings = []
if not has_permit:
findings.append("Missing confined space entry permit")
if not has_attendant:
findings.append("No attendant stationed at entry")
if not atmospheric_tested:
findings.append("Atmospheric testing not performed")
checks.append(ComplianceCheck(
rule_id=cs_rule.rule_id,
rule_name=cs_rule.name,
status=ComplianceStatus.COMPLIANT if not findings else ComplianceStatus.NON_COMPLIANT,
findings=findings,
evidence=f"Confined space: {location}",
checked_at=datetime.now(),
checked_by="automated_system"
))
return checks
def create_violation(self, check: ComplianceCheck, location: str) -> Violation:
"""Create violation record from failed check"""
rule = self.rules.get(check.rule_id)
# Determine deadline based on severity
if rule.risk_level == RiskLevel.CRITICAL:
deadline = date.today() # Immediate
elif rule.risk_level == RiskLevel.HIGH:
deadline = date.today() + timedelta(days=1)
elif rule.risk_level == RiskLevel.MEDIUM:
deadline = date.today() + timedelta(days=3)
else:
deadline = date.today() + timedelta(days=7)
violation = Violation(
rule=rule,
location=location,
description="; ".join(check.findings),
detected_at=check.checked_at,
severity=rule.risk_level,
corrective_action=self._get_corrective_action(rule, check.findings),
deadline=deadline
)
self.violations.append(violation)
return violation
def _get_corrective_action(self, rule: SafetyRule, findings: List[str]) -> str:
"""Generate corrective action based on rule and findings"""
actions = {
RuleCategory.PPE: "Provide required PPE to workers and ensure proper usage",
RuleCategory.FALL_PROTECTION: "Install guardrails, safety nets, or provide personal fall arrest systems",
RuleCategory.ELECTRICAL: "De-energize equipment, install GFCI protection, verify grounding",
RuleCategory.EXCAVATION: "Perform required inspections, install shoring/sloping as needed",
RuleCategory.CONFINED_SPACE: "Stop work, evacuate space, obtain required permits",
RuleCategory.HOT_WORK: "Obtain hot work permit, assign fire watch, provide extinguishers",
RuleCategory.SCAFFOLDING: "Inspect scaffold, install missing components, tag out if unsafe",
RuleCategory.CRANE: "Verify operator certification, check rigging, establish swing radius"
}
return actions.get(rule.category, "Review and correct violation immediately")
class PPEComplianceChecker:
"""PPE compliance checking with CV integration support"""
def __init__(self):
self.ppe_requirements = self._load_requirements()
def _load_requirements(self) -> Dict[str, List[str]]:
"""Load PPE requirements by work type"""
return {
'general': ['helmet', 'safety_vest', 'safety_boots'],
'welding': ['helmet', 'welding_mask', 'gloves', 'apron', 'safety_boots'],
'excavation': ['helmet', 'safety_vest', 'safety_boots', 'gloves'],
'electrical': ['helmet', 'safety_glasses', 'insulated_gloves', 'safety_boots'],
'concrete': ['helmet', 'safety_glasses', 'gloves', 'safety_boots', 'knee_pads'],
'demolition': ['helmet', 'safety_glasses', 'dust_mask', 'gloves', 'safety_boots'],
'roofing': ['helmet', 'harness', 'safety_boots', 'gloves'],
'painting': ['helmet', 'respirator', 'safety_glasses', 'gloves', 'coveralls'],
'scaffolding': ['helmet', 'harness', 'safety_boots', 'gloves']
}
def check_worker_ppe(self, worker_id: str, work_type: str,
detected_ppe: List[str]) -> Dict:
"""Check worker PPE against requirements"""
required = self.ppe_requirements.get(work_type, self.ppe_requirements['general'])
missing = set(required) - set(detected_ppe)
extra = set(detected_ppe) - set(required)
compliance_score = (len(required) - len(missing)) / len(required) * 100
return {
'worker_id': worker_id,
'work_type': work_type,
'required_ppe': required,
'detected_ppe': detected_ppe,
'missing_ppe': list(missing),
'compliance_score': compliance_score,
'is_compliant': len(missing) == 0,
'recommendations': [f"Provide {item}" for item in missing]
}
def process_cv_detections(self, detections: List[Dict]) -> List[Dict]:
"""Process detections from computer vision system"""
results = []
for detection in detections:
worker_id = detection.get('worker_id', 'unknown')
work_type = detection.get('work_type', 'general')
detected_items = detection.get('ppe_items', [])
result = self.check_worker_ppe(worker_id, work_type, detected_items)
result['timestamp'] = detection.get('timestamp')
result['location'] = detection.get('location')
result['image_ref'] = detection.get('image_path')
results.append(result)
return results
from datetime import datetime, date
from typing import List, Dict, Optional
import pandas as pd
@dataclass
class InspectionItem:
item_id: str
category: str
description: str
is_compliant: bool
notes: str
photo_refs: List[str] = field(default_factory=list)
corrective_action: Optional[str] = None
@dataclass
class SiteInspection:
inspection_id: str
site_id: str
inspector: str
inspection_date: date
weather: str
items: List[InspectionItem]
overall_score: float
recommendations: List[str]
class SiteInspectionSystem:
"""Construction site safety inspection management"""
def __init__(self):
self.checklists = self._load_checklists()
self.inspections: List[SiteInspection] = []
def _load_checklists(self) -> Dict[str, List[Dict]]:
"""Load inspection checklists by category"""
return {
'general_safety': [
{'id': 'GS-001', 'item': 'Site access controlled and secured', 'category': 'Access'},
{'id': 'GS-002', 'item': 'Safety signage posted at entrance', 'category': 'Signage'},
{'id': 'GS-003', 'item': 'First aid station accessible and stocked', 'category': 'Emergency'},
{'id': 'GS-004', 'item': 'Emergency contact numbers posted', 'category': 'Emergency'},
{'id': 'GS-005', 'item': 'Fire extinguishers accessible and inspected', 'category': 'Fire'},
{'id': 'GS-006', 'item': 'Evacuation routes clear and marked', 'category': 'Emergency'},
{'id': 'GS-007', 'item': 'Housekeeping adequate (no tripping hazards)', 'category': 'Housekeeping'},
{'id': 'GS-008', 'item': 'Material storage organized and stable', 'category': 'Storage'},
{'id': 'GS-009', 'item': 'Adequate lighting in work areas', 'category': 'Environment'},
{'id': 'GS-010', 'item': 'Toilets and washing facilities clean', 'category': 'Welfare'}
],
'fall_protection': [
{'id': 'FP-001', 'item': 'Guardrails installed where required', 'category': 'Guardrails'},
{'id': 'FP-002', 'item': 'Floor openings covered or protected', 'category': 'Openings'},
{'id': 'FP-003', 'item': 'Ladders in good condition and secured', 'category': 'Ladders'},
{'id': 'FP-004', 'item': 'Harnesses inspected and in good condition', 'category': 'PPE'},
{'id': 'FP-005', 'item': 'Anchor points adequate and tested', 'category': 'Anchors'},
{'id': 'FP-006', 'item': 'Safety nets installed where required', 'category': 'Nets'},
{'id': 'FP-007', 'item': 'Stairways have handrails', 'category': 'Stairs'},
{'id': 'FP-008', 'item': 'Roof edge protection in place', 'category': 'Roof'}
],
'scaffolding': [
{'id': 'SC-001', 'item': 'Scaffold erected by competent person', 'category': 'Erection'},
{'id': 'SC-002', 'item': 'Base plates and mudsills in place', 'category': 'Foundation'},
{'id': 'SC-003', 'item': 'All platforms fully planked', 'category': 'Platforms'},
{'id': 'SC-004', 'item': 'Guardrails on all open sides', 'category': 'Protection'},
{'id': 'SC-005', 'item': 'Access ladders provided', 'category': 'Access'},
{'id': 'SC-006', 'item': 'Tied to structure at required intervals', 'category': 'Stability'},
{'id': 'SC-007', 'item': 'Inspection tag current', 'category': 'Documentation'},
{'id': 'SC-008', 'item': 'No damage to components', 'category': 'Condition'}
],
'electrical': [
{'id': 'EL-001', 'item': 'GFCI protection on all outlets', 'category': 'Protection'},
{'id': 'EL-002', 'item': 'Extension cords in good condition', 'category': 'Equipment'},
{'id': 'EL-003', 'item': 'Panel boxes closed and labeled', 'category': 'Panels'},
{'id': 'EL-004', 'item': 'No exposed live wires', 'category': 'Hazards'},
{'id': 'EL-005', 'item': 'Lockout/tagout procedures followed', 'category': 'LOTO'},
{'id': 'EL-006', 'item': 'Temporary wiring properly supported', 'category': 'Installation'}
],
'excavation': [
{'id': 'EX-001', 'item': 'Competent person inspection completed', 'category': 'Inspection'},
{'id': 'EX-002', 'item': 'Utilities located and marked', 'category': 'Utilities'},
{'id': 'EX-003', 'item': 'Protective systems in place (shoring/sloping)', 'category': 'Protection'},
{'id': 'EX-004', 'item': 'Spoil pile minimum 2ft from edge', 'category': 'Materials'},
{'id': 'EX-005', 'item': 'Safe access/egress within 25ft', 'category': 'Access'},
{'id': 'EX-006', 'item': 'Water accumulation managed', 'category': 'Water'},
{'id': 'EX-007', 'item': 'Barricades around excavation', 'category': 'Barriers'}
]
}
def conduct_inspection(self, site_id: str, inspector: str,
checklist_types: List[str],
responses: Dict[str, Dict]) -> SiteInspection:
"""Conduct site inspection"""
items = []
for checklist_type in checklist_types:
checklist = self.checklists.get(checklist_type, [])
for check in checklist:
item_id = check['id']
response = responses.get(item_id, {})
items.append(InspectionItem(
item_id=item_id,
category=check['category'],
description=check['item'],
is_compliant=response.get('compliant', False),
notes=response.get('notes', ''),
photo_refs=response.get('photos', []),
corrective_action=response.get('action') if not response.get('compliant', False) else None
))
# Calculate score
compliant_count = sum(1 for item in items if item.is_compliant)
overall_score = (compliant_count / len(items) * 100) if items else 0
# Generate recommendations
recommendations = []
for item in items:
if not item.is_compliant:
recommendations.append(f"{item.item_id}: {item.corrective_action or 'Address non-compliance'}")
inspection = SiteInspection(
inspection_id=f"INS-{site_id}-{datetime.now().strftime('%Y%m%d%H%M')}",
site_id=site_id,
inspector=inspector,
inspection_date=date.today(),
weather="Clear", # Would be input
items=items,
overall_score=overall_score,
recommendations=recommendations[:10] # Top 10
)
self.inspections.append(inspection)
return inspection
def generate_inspection_report(self, inspection: SiteInspection) -> pd.DataFrame:
"""Generate inspection report DataFrame"""
data = []
for item in inspection.items:
data.append({
'Inspection ID': inspection.inspection_id,
'Date': inspection.inspection_date,
'Inspector': inspection.inspector,
'Item ID': item.item_id,
'Category': item.category,
'Description': item.description,
'Compliant': 'Yes' if item.is_compliant else 'No',
'Notes': item.notes,
'Corrective Action': item.corrective_action or ''
})
return pd.DataFrame(data)
def get_compliance_trends(self, site_id: str, days: int = 30) -> Dict:
"""Get compliance trends for a site"""
site_inspections = [
i for i in self.inspections
if i.site_id == site_id and
(date.today() - i.inspection_date).days <= days
]
if not site_inspections:
return {'message': 'No inspections found'}
scores = [i.overall_score for i in site_inspections]
dates = [i.inspection_date for i in site_inspections]
# Category breakdown
category_compliance = {}
for inspection in site_inspections:
for item in inspection.items:
if item.category not in category_compliance:
category_compliance[item.category] = {'compliant': 0, 'total': 0}
category_compliance[item.category]['total'] += 1
if item.is_compliant:
category_compliance[item.category]['compliant'] += 1
return {
'site_id': site_id,
'inspection_count': len(site_inspections),
'average_score': sum(scores) / len(scores),
'score_trend': list(zip(dates, scores)),
'category_compliance': {
k: v['compliant'] / v['total'] * 100
for k, v in category_compliance.items()
},
'latest_inspection': site_inspections[-1].inspection_id
}
class SafetyDashboard:
"""Generate safety compliance dashboard data"""
def __init__(self, engine: SafetyComplianceEngine,
inspection_system: SiteInspectionSystem):
self.engine = engine
self.inspections = inspection_system
def get_dashboard_data(self, site_id: str) -> Dict:
"""Get comprehensive dashboard data"""
# Active violations
open_violations = [v for v in self.engine.violations if v.status == 'open']
# Violations by severity
by_severity = {}
for v in open_violations:
sev = v.severity.name
by_severity[sev] = by_severity.get(sev, 0) + 1
# Violations by category
by_category = {}
for v in open_violations:
cat = v.rule.category.value
by_category[cat] = by_category.get(cat, 0) + 1
# Overdue violations
overdue = [v for v in open_violations if v.deadline < date.today()]
# Compliance trends
trends = self.inspections.get_compliance_trends(site_id)
return {
'summary': {
'total_open_violations': len(open_violations),
'critical_violations': by_severity.get('CRITICAL', 0),
'overdue_violations': len(overdue),
'latest_inspection_score': trends.get('average_score', 0)
},
'violations_by_severity': by_severity,
'violations_by_category': by_category,
'overdue_items': [
{
'rule': v.rule.name,
'location': v.location,
'deadline': v.deadline.isoformat(),
'days_overdue': (date.today() - v.deadline).days
}
for v in overdue
],
'compliance_trend': trends.get('score_trend', []),
'category_compliance': trends.get('category_compliance', {})
}
| Check Type | OSHA Reference | Risk Level | Frequency |
|------------|---------------|------------|-----------|
| PPE Compliance | 1926.100-106 | High | Continuous |
| Fall Protection | 1926.501-503 | Critical | Daily |
| Scaffolding | 1926.451-454 | High | Before each use |
| Excavation | 1926.651-652 | High | Daily |
| Electrical | 1926.400-449 | Critical | Daily |
| Confined Space | 1926.1200 | Critical | Before entry |
| Hot Work | 1926.350-354 | High | Per activity |
progress-monitoring-cv for PPE detection with computer visionrisk-assessment-ml for predictive safety analyticsdocument-classification-nlp for safety document processingComprehensive spreadsheet creation, editing, and analysis with support for formulas, formatting, data analysis, and visualization. When Claude needs to work with spreadsheets (.xlsx, .xlsm, .csv, .tsv, etc) for: (1) Creating new spreadsheets with formulas and formatting, (2) Reading or analyzing data, (3) Modify existing spreadsheets while preserving formulas, (4) Data analysis and visualization in spreadsheets, or (5) Recalculating formulas
Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.
Picks random winners from lists, spreadsheets, or Google Sheets for giveaways, raffles, and contests. Ensures fair, unbiased selection with transparency.
Query openFDA API for drugs, devices, adverse events, recalls, regulatory submissions (510k, PMA), substance identification (UNII), for FDA regulatory data analysis and safety research.
MATLAB and GNU Octave numerical computing for matrix operations, data analysis, visualization, and scientific computing. Use when writing MATLAB/Octave scripts for linear algebra, signal processing, image processing, differential equations, optimization, statistics, or creating scientific visualizations. Also use when the user needs help with MATLAB syntax, functions, or wants to convert between MATLAB and Python code. Scripts can be executed with MATLAB or the open-source GNU Octave interpreter.
UMAP dimensionality reduction. Fast nonlinear manifold learning for 2D/3D visualization, clustering preprocessing (HDBSCAN), supervised/parametric UMAP, for high-dimensional data.
Creating interactive data visualisations using d3.js. This skill should be used when creating custom charts, graphs, network diagrams, geographic visualisations, or any complex SVG-based data visualisation that requires fine-grained control over visual elements, transitions, or interactions. Use this for bespoke visualisations beyond standard charting libraries, whether in React, Vue, Svelte, vanilla JavaScript, or any other environment.
Access AlphaFold 200M+ AI-predicted protein structures. Retrieve structures by UniProt ID, download PDB/mmCIF files, analyze confidence metrics (pLDDT, PAE), for drug discovery and structural biology.
Take datadrivenconstruction/safety-compliance-checker from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.