daily-report-generator — скилл для Claude Code: автоматизация отчётов
Скилл daily-report-generator собирает данные со строительной площадки — погоду, рабочие часы, прогресс — и генерирует профессиональные PDF-отчёты. Экономит 45–60 минут в день на ручное составление документов.
MIT 130 установок ★ 434 обновлён 2026-10-19
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill daily-report-generator -a claude-codeЧто делает daily-report-generator
Скилл интегрирует данные из Google Sheets и баз проектов, подтягивает погоду через API, агрегирует табели рабочих и создаёт единый PDF с таблицами, фото и выводами. Внутри — Python с ReportLab для вёрстки отчётов и OpenWeatherMap для метеоданных.
Отчёт включает погодные условия, численность и часы по специальностям, выполненные и планируемые работы, проблемы, инциденты и фотографии. Распределяет готовый документ заказчику, архитектору и PM автоматически.
Требует Python 3 в PATH. Для работы с погодой нужен API-ключ OpenWeatherMap, для интеграции с источниками данных — доступ к Google Sheets или проектной базе.
Какие задачи решает
- собрать дневной отчёт с участка за один клик
- вставить погоду, табели и фото в единый PDF
- распределить готовый документ подрядчику и архитектору
- сократить время на ручное составление отчётов
- сохранить историю работ в едином формате
Примеры запросов агенту
- Создай дневной отчёт по участку за 24 января с фото из папки
photos - Соберёшь отчёт по электромонтажу и сантехнике с погодой на Москву?
- Сгенерируй PDF с табелем рабочих, прогрессом и проблемами дня
Как установить daily-report-generator
1Через Skills CLI (Vercel)
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill daily-report-generator -a claude-code2Вручную: положить файлы
~/.claude/skills/daily-report-generator/SKILL.mdПапка скилла целиком; для одного проекта — .claude/skills/daily-report-generator/
1Через Skills CLI (Vercel)
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill daily-report-generator -a codex2Вручную: положить файлы
.agents/skills/daily-report-generator/SKILL.mdДля всех проектов — ~/.agents/skills/
1Через Skills CLI (Vercel)
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill daily-report-generator -a cursor2Вручную: положить файлы
.cursor/skills/daily-report-generator/SKILL.mdCursor читает и .agents/skills/, и .claude/skills/
1GitHub CLI
gh skill install datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction 1_DDC_Toolkit/Field-Operations/daily-report-generator2Вручную: положить файлы
.github/skills/daily-report-generator/SKILL.md1Через Skills CLI (Vercel)
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill daily-report-generator -a gemini-cli2Командой платформы
gemini skills install https://github.com/datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --path 1_DDC_Toolkit/Field-Operations/daily-report-generatorВручную: положить файлы
.devin/skills/daily-report-generator/SKILL.mdСтарый путь .windsurf/skills/ тоже читается
Через Skills CLI (Vercel)
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill daily-report-generator -a opencodeClaude.ai
Упакуйте папку в zip и загрузите в Settings → Capabilities → Skills
По шагам, с командами для каждой платформы: Как установить скилл в Claude Code: пять способов
Вопросы о daily-report-generator
Как установить daily-report-generator в Claude Code?
Выполни в терминале: npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill daily-report-generator -a claude-code, затем открой ~/.claude/skills/daily-report-generator/SKILL.md.
Что делает daily-report-generator?
Автоматизирует дневные отчёты строительной площадки: собирает данные о рабочих, погоде и прогрессе, вставляет фото и создаёт профессиональный PDF за несколько минут вместо часа ручной работы.
Какие данные нужны для работы скилла?
Python 3, API-ключ OpenWeatherMap для погоды, доступ к Google Sheets или базе проекта, фотографии в папке. Если ключа нет, скилл использует тестовые данные.
Исходный текст
SKILL.md · MIT
---
name: "daily-report-generator"
description: "Automatically generate daily construction reports from field data, worker inputs, weather, and progress photos. Creates professional PDF reports."
homepage: "https://datadrivenconstruction.io"
metadata: {"openclaw": {"emoji": "🏗️", "os": ["darwin", "linux", "win32"], "homepage": "https://datadrivenconstruction.io", "requires": {"bins": ["python3"]}}}
---
# Daily Report Generator for Construction Sites
Automate the creation of comprehensive daily construction reports by aggregating data from multiple sources into professional documentation.
## Business Case
**Problem**: Site managers spend 45-60 minutes daily on:
- Collecting information from foremen
- Checking weather conditions
- Compiling worker counts and hours
- Writing narrative summaries
- Formatting and distributing reports
**Solution**: Automated system that:
- Pulls data from Google Sheets/project database
- Integrates weather API data
- Aggregates worker timesheets
- Generates professional PDF reports
- Distributes to stakeholders automatically
**ROI**: 80% reduction in daily reporting time (45 min → 9 min for review)
## Report Structure
```
┌──────────────────────────────────────────────────────────────────────┐
│ DAILY CONSTRUCTION REPORT │
│ │
│ Project: ЖК Солнечный, Корпус 2 Date: 24.01.2026 │
│ Report #: DCR-2026-024 Weather: ☁️ -5°C │
├──────────────────────────────────────────────────────────────────────┤
│ │
│ 1. WEATHER CONDITIONS │
│ ┌────────────┬────────────┬────────────┬────────────┐ │
│ │ Morning │ Afternoon │ Evening │ Impact │ │
│ │ -8°C ☀️ │ -5°C ☁️ │ -7°C 🌙 │ Normal │ │
│ └────────────┴────────────┴────────────┴────────────┘ │
│ │
│ 2. WORKFORCE │
│ ┌────────────────────────────────────────────────────┐ │
│ │ Category │ Planned │ Actual │ Hours │ │
│ ├────────────────────────────────────────────────────┤ │
│ │ GC Supervision │ 3 │ 3 │ 27 │ │
│ │ Electrical │ 12 │ 11 │ 88 │ │
│ │ Plumbing │ 8 │ 8 │ 64 │ │
│ │ HVAC │ 6 │ 6 │ 48 │ │
│ │ TOTAL │ 29 │ 28 │ 227 │ │
│ └────────────────────────────────────────────────────┘ │
│ │
│ 3. WORK COMPLETED TODAY │
│ • Electrical: Completed rough-in floors 5-6 │
│ • Plumbing: Installed risers section A │
│ • HVAC: Ductwork installation 60% complete │
│ │
│ 4. WORK PLANNED FOR TOMORROW │
│ • Electrical: Begin rough-in floor 7 │
│ • Plumbing: Continue risers section B │
│ • HVAC: Complete ductwork, begin testing │
│ │
│ 5. ISSUES / DELAYS │
│ • Material delay: Electrical panels (ETA: 26.01) │
│ • Weather: Expected snow may delay exterior work │
│ │
│ 6. SAFETY │
│ ✅ No incidents │
│ ✅ Toolbox talk completed: Fall protection │
│ │
│ 7. PHOTOS │
│ [Photo 1: Floor 5 electrical] [Photo 2: Riser installation] │
│ │
│ ───────────────────────────────────────────────────────────────── │
│ Prepared by: Иван Петров, Site Manager │
│ Approved by: ___________________ │
│ Distribution: Owner, Architect, PM │
└──────────────────────────────────────────────────────────────────────┘
```
## Python Implementation
```python
import pandas as pd
from datetime import datetime, date
from typing import Optional, List, Dict
import requests
from reportlab.lib import colors
from reportlab.lib.pagesizes import A4
from reportlab.platypus import SimpleDocTemplate, Table, TableStyle, Paragraph, Spacer, Image
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
from reportlab.lib.units import cm
import os
class DailyReportGenerator:
"""Generate professional daily construction reports"""
def __init__(self, config: dict):
self.config = config
self.weather_api_key = config.get('weather_api_key')
self.project_name = config.get('project_name')
self.report_date = config.get('report_date', date.today())
def get_weather_data(self, location: str) -> dict:
"""Fetch weather data from API"""
if not self.weather_api_key:
return self._mock_weather()
url = f"https://api.openweathermap.org/data/2.5/weather"
params = {
'q': location,
'appid': self.weather_api_key,
'units': 'metric',
'lang': 'ru'
}
response = requests.get(url, params=params)
if response.status_code == 200:
data = response.json()
return {
'temp': round(data['main']['temp']),
'description': data['weather'][0]['description'],
'humidity': data['main']['humidity'],
'wind_speed': round(data['wind']['speed']),
'icon': self._get_weather_icon(data['weather'][0]['main'])
}
return self._mock_weather()
def _get_weather_icon(self, condition: str) -> str:
icons = {
'Clear': '☀️',
'Clouds': '☁️',
'Rain': '🌧️',
'Snow': '❄️',
'Thunderstorm': '⛈️',
'Mist': '🌫️'
}
return icons.get(condition, '🌤️')
def _mock_weather(self) -> dict:
return {
'temp': -5,
'description': 'облачно',
'humidity': 65,
'wind_speed': 3,
'icon': '☁️'
}
def get_workforce_data(self, source: pd.DataFrame) -> dict:
"""Aggregate workforce data from timesheet"""
# Expected columns: trade, worker_name, hours_worked, planned_hours
summary = source.groupby('trade').agg({
'worker_name': 'count',
'hours_worked': 'sum',
'planned_hours': 'sum'
}).reset_index()
summary.columns = ['trade', 'actual_count', 'actual_hours', 'planned_hours']
# Calculate planned count (assuming 8-hour shifts)
summary['planned_count'] = (summary['planned_hours'] / 8).astype(int)
return {
'trades': summary.to_dict('records'),
'total_workers': summary['actual_count'].sum(),
'total_hours': summary['actual_hours'].sum(),
'total_planned': summary['planned_count'].sum()
}
def get_work_completed(self, tasks: pd.DataFrame) -> List[dict]:
"""Extract completed work from task system"""
# Filter completed tasks for today
completed = tasks[
(tasks['date'] == self.report_date.strftime('%d.%m.%Y')) &
(tasks['status'].isin(['Completed', 'Partial']))
]
work_items = []
for _, row in completed.iterrows():
work_items.append({
'trade': row['trade'],
'description': row['description'],
'status': row['status'],
'notes': row.get('notes', '')
})
return work_items
def get_work_planned(self, tasks: pd.DataFrame) -> List[dict]:
"""Get planned work for tomorrow"""
tomorrow = self.report_date + pd.Timedelta(days=1)
planned = tasks[
tasks['date'] == tomorrow.strftime('%d.%m.%Y')
]
work_items = []
for _, row in planned.iterrows():
work_items.append({
'trade': row['trade'],
'description': row['description'],
'priority': row.get('priority', 'Medium')
})
return work_items
def get_issues(self, issues_log: pd.DataFrame) -> List[dict]:
"""Get active issues and delays"""
active = issues_log[
(issues_log['status'] == 'Open') |
(issues_log['date_reported'] == self.report_date.strftime('%d.%m.%Y'))
]
return active[['category', 'description', 'impact', 'resolution_date']].to_dict('records')
def get_safety_data(self, safety_log: pd.DataFrame) -> dict:
"""Get safety information for the day"""
today_incidents = safety_log[
safety_log['date'] == self.report_date.strftime('%d.%m.%Y')
]
return {
'incidents': len(today_incidents[today_incidents['type'] == 'Incident']),
'near_misses': len(today_incidents[today_incidents['type'] == 'Near Miss']),
'toolbox_talk': today_incidents[
today_incidents['type'] == 'Toolbox Talk'
]['topic'].tolist(),
'observations': today_incidents[
today_incidents['type'] == 'Observation'
]['description'].tolist()
}
def generate_report(self, data: dict, output_path: str) -> str:
"""Generate PDF report"""
doc = SimpleDocTemplate(
output_path,
pagesize=A4,
rightMargin=2*cm,
leftMargin=2*cm,
topMargin=2*cm,
bottomMargin=2*cm
)
styles = getSampleStyleSheet()
title_style = ParagraphStyle(
'Title',
parent=styles['Heading1'],
fontSize=16,
alignment=1,
spaceAfter=12
)
heading_style = ParagraphStyle(
'Heading',
parent=styles['Heading2'],
fontSize=12,
spaceBefore=12,
spaceAfter=6
)
elements = []
# Title
elements.append(Paragraph(
f"DAILY CONSTRUCTION REPORT",
title_style
))
# Header info
header_data = [
['Project:', self.project_name, 'Date:', self.report_date.strftime('%d.%m.%Y')],
['Report #:', data.get('report_number', 'DCR-001'), 'Weather:', f"{data['weather']['icon']} {data['weather']['temp']}°C"]
]
header_table = Table(header_data, colWidths=[3*cm, 6*cm, 3*cm, 4*cm])
header_table.setStyle(TableStyle([
('FONTNAME', (0, 0), (-1, -1), 'Helvetica'),
('FONTSIZE', (0, 0), (-1, -1), 10),
('FONTNAME', (0, 0), (0, -1), 'Helvetica-Bold'),
('FONTNAME', (2, 0), (2, -1), 'Helvetica-Bold'),
]))
elements.append(header_table)
elements.append(Spacer(1, 12))
# Weather section
elements.append(Paragraph("1. WEATHER CONDITIONS", heading_style))
weather = data['weather']
weather_text = f"""
Temperature: {weather['temp']}°C | Humidity: {weather['humidity']}% |
Wind: {weather['wind_speed']} m/s | Conditions: {weather['description']}
"""
elements.append(Paragraph(weather_text, styles['Normal']))
# Workforce section
elements.append(Paragraph("2. WORKFORCE", heading_style))
workforce = data['workforce']
workforce_data = [['Trade', 'Planned', 'Actual', 'Hours']]
for trade in workforce['trades']:
workforce_data.append([
trade['trade'],
str(trade['planned_count']),
str(trade['actual_count']),
str(int(trade['actual_hours']))
])
workforce_data.append([
'TOTAL',
str(workforce['total_planned']),
str(workforce['total_workers']),
str(int(workforce['total_hours']))
])
workforce_table = Table(workforce_data, colWidths=[6*cm, 3*cm, 3*cm, 3*cm])
workforce_table.setStyle(TableStyle([
('BACKGROUND', (0, 0), (-1, 0), colors.grey),
('TEXTCOLOR', (0, 0), (-1, 0), colors.whitesmoke),
('FONTNAME', (0, 0), (-1, 0), 'Helvetica-Bold'),
('FONTNAME', (0, -1), (-1, -1), 'Helvetica-Bold'),
('GRID', (0, 0), (-1, -1), 1, colors.black),
('ALIGN', (1, 0), (-1, -1), 'CENTER'),
]))
elements.append(workforce_table)
# Work completed
elements.append(Paragraph("3. WORK COMPLETED TODAY", heading_style))
for item in data.get('work_completed', []):
bullet = f"• {item['trade']}: {item['description']}"
if item.get('notes'):
bullet += f" ({item['notes']})"
elements.append(Paragraph(bullet, styles['Normal']))
# Work planned
elements.append(Paragraph("4. WORK PLANNED FOR TOMORROW", heading_style))
for item in data.get('work_planned', []):
bullet = f"• {item['trade']}: {item['description']}"
elements.append(Paragraph(bullet, styles['Normal']))
# Issues
elements.append(Paragraph("5. ISSUES / DELAYS", heading_style))
issues = data.get('issues', [])
if issues:
for issue in issues:
bullet = f"• {issue['category']}: {issue['description']}"
if issue.get('resolution_date'):
bullet += f" (ETA: {issue['resolution_date']})"
elements.append(Paragraph(bullet, styles['Normal']))
else:
elements.append(Paragraph("No significant issues reported.", styles['Normal']))
# Safety
elements.append(Paragraph("6. SAFETY", heading_style))
safety = data.get('safety', {})
if safety.get('incidents', 0) == 0:
elements.append(Paragraph("✅ No incidents reported", styles['Normal']))
else:
elements.append(Paragraph(f"⚠️ {safety['incidents']} incident(s) reported", styles['Normal']))
if safety.get('toolbox_talk'):
elements.append(Paragraph(f"✅ Toolbox talk: {', '.join(safety['toolbox_talk'])}", styles['Normal']))
# Signature block
elements.append(Spacer(1, 24))
elements.append(Paragraph("─" * 60, styles['Normal']))
elements.append(Paragraph(f"Prepared by: {data.get('prepared_by', '_________________')}", styles['Normal']))
elements.append(Paragraph(f"Date: {datetime.now().strftime('%d.%m.%Y %H:%M')}", styles['Normal']))
# Build PDF
doc.build(elements)
return output_path
# Usage Example
def generate_daily_report(
project_name: str,
location: str,
timesheet_path: str,
tasks_path: str,
output_dir: str
) -> str:
"""Generate daily report from source files"""
# Initialize generator
generator = DailyReportGenerator({
'project_name': project_name,
'weather_api_key': os.environ.get('WEATHER_API_KEY'),
'report_date': date.today()
})
# Load data
timesheet = pd.read_excel(timesheet_path)
tasks = pd.read_excel(tasks_path)
# Compile report data
report_data = {
'report_number': f"DCR-{date.today().strftime('%Y-%j')}",
'weather': generator.get_weather_data(location),
'workforce': generator.get_workforce_data(timesheet),
'work_completed': generator.get_work_completed(tasks),
'work_planned': generator.get_work_planned(tasks),
'issues': [], # Load from issues log if available
'safety': {
'incidents': 0,
'toolbox_talk': ['Fall Protection'],
'near_misses': 0
},
'prepared_by': 'Site Manager'
}
# Generate PDF
output_path = os.path.join(
output_dir,
f"Daily_Report_{date.today().strftime('%Y%m%d')}.pdf"
)
return generator.generate_report(report_data, output_path)
if __name__ == "__main__":
report_path = generate_daily_report(
project_name="ЖК Солнечный, Корпус 2",
location="Moscow,RU",
timesheet_path="timesheet.xlsx",
tasks_path="tasks.xlsx",
output_dir="./reports"
)
print(f"Report generated: {report_path}")
```
## Data Sources Integration
### From n8n Project Management System
```python
# Connect to Google Sheets used by n8n bot
def get_data_from_project_management(spreadsheet_id: str) -> dict:
"""Pull data from n8n project management system"""
import gspread
gc = gspread.service_account()
sh = gc.open_by_key(spreadsheet_id)
# Get completed tasks
tasks_sheet = sh.worksheet('Tasks')
tasks = pd.DataFrame(tasks_sheet.get_all_records())
# Get workforce from worker responses
workers_sheet = sh.worksheet('Workers')
workers = pd.DataFrame(workers_sheet.get_all_records())
return {
'tasks': tasks,
'workers': workers
}
```
### From Timesheet System
```python
# Integrate with common timesheet formats
def import_timesheet(source: str, format: str = 'excel') -> pd.DataFrame:
"""Import timesheet data from various sources"""
if format == 'excel':
df = pd.read_excel(source)
elif format == 'csv':
df = pd.read_csv(source)
elif format == 'procore':
df = fetch_procore_timesheet(source)
# Standardize columns
df = df.rename(columns={
'Trade': 'trade',
'Worker': 'worker_name',
'Hours': 'hours_worked',
'Planned Hours': 'planned_hours'
})
return df
```
## n8n Workflow for Automation
```yaml
name: Daily Report Automation
trigger:
type: cron
expression: "0 18 * * 1-6" # 6 PM daily
steps:
- collect_task_data:
node: Google Sheets
operation: readRows
sheet: Tasks
filter: Date = TODAY()
- collect_timesheet:
node: Google Sheets
operation: readRows
sheet: Timesheet
filter: Date = TODAY()
- get_weather:
node: HTTP Request
url: "https://api.openweathermap.org/data/2.5/weather"
params:
q: "Moscow,RU"
appid: "{{$env.WEATHER_API_KEY}}"
- generate_report:
node: Code (Python)
code: |
from daily_report import generate_report
return generate_report(items)
- upload_to_drive:
node: Google Drive
operation: upload
file: "={{$json.report_path}}"
folder: "Daily Reports"
- send_notification:
node: Telegram
operation: sendDocument
chatId: "MANAGERS_GROUP_ID"
document: "={{$json.drive_url}}"
caption: |
📋 Daily Report - {{$now.format('DD.MM.YYYY')}}
Workforce: {{$json.total_workers}} workers
Tasks completed: {{$json.completed_tasks}}
Issues: {{$json.open_issues}}
```
## Report Distribution
```python
def distribute_report(report_path: str, recipients: dict):
"""Distribute report to stakeholders"""
# Email distribution
for email in recipients.get('email', []):
send_email(
to=email,
subject=f"Daily Report - {date.today().strftime('%d.%m.%Y')}",
body="Please find attached the daily construction report.",
attachment=report_path
)
# Telegram distribution
for chat_id in recipients.get('telegram', []):
send_telegram_document(
chat_id=chat_id,
document_path=report_path,
caption=f"📋 Daily Report - {date.today().strftime('%d.%m.%Y')}"
)
# Upload to project portal
if portal_url := recipients.get('portal'):
upload_to_portal(portal_url, report_path)
```
## Best Practices
1. **Data Collection**: Set up automated data collection to minimize manual input
2. **Review Time**: Allow 5-10 minutes for manager review before distribution
3. **Photos**: Include 3-5 key photos showing progress
4. **Issues**: Be specific about impacts and resolution dates
5. **Distribution**: Send by 6-7 PM to allow stakeholder review
---
*"A good daily report tells the story of the day in 2 minutes or less."*
© datadrivenconstruction · MIT · datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction