name: report-generator description: 从数据中生成专业的markdown和HTML报告，包括图表、表格和分析。
报告生成器技能
从数据中生成专业的markdown和HTML报告，包括图表、表格和分析。
指令
您是一个报告生成专家。当调用时：
-
分析数据：
- 理解数据结构和内容
- 识别关键指标和见解
- 计算统计和趋势
- 检测模式和异常
- 生成执行摘要
-
创建报告结构：
- 设计清晰、逻辑的部分
- 创建目录
- 添加执行摘要
- 包括详细分析
- 提供建议
-
生成可视化：
- 为结构化数据创建表格
- 生成图表（条形图、折线图、饼图、散点图）
- 添加徽章和指标
- 包括代码块和示例
- 格式化数字和百分比
-
格式化输出：
- 生成markdown报告
- 创建带有样式的HTML报告
- 导出到PDF
- 添加品牌和定制
- 确保响应式设计
使用示例
@report-generator data.csv
@report-generator --format html
@report-generator --template executive-summary
@report-generator --charts --pdf
@report-generator --compare baseline.json current.json
报告类型
执行摘要报告
def generate_executive_summary(data, title="执行摘要"):
"""
生成高级执行摘要报告
"""
from datetime import datetime
report = f"""# {title}
**生成时间:** {datetime.now().strftime('%B %d, %Y at %I:%M %p')}
---
## 关键亮点
"""
# 计算关键指标
metrics = calculate_key_metrics(data)
for metric in metrics:
icon = "✅" if metric['status'] == 'good' else "⚠️" if metric['status'] == 'warning' else "❌"
report += f"{icon} **{metric['name']}**: {metric['value']}
"
report += f"""
---
## 性能概览
| 指标 | 当前 | 之前 | 变化 |
|--------|---------|----------|--------|
"""
for metric in metrics:
if 'previous' in metric:
change = calculate_change(metric['current'], metric['previous'])
arrow = "↑" if change > 0 else "↓" if change < 0 else "→"
color = "green" if change > 0 else "red" if change < 0 else "gray"
report += f"| {metric['name']} | {metric['current']:,} | {metric['previous']:,} | {arrow} {abs(change):.1f}% |
"
report += """
---
## 建议
"""
recommendations = generate_recommendations(metrics)
for i, rec in enumerate(recommendations, 1):
priority = rec.get('priority', 'medium')
emoji = "🔴" if priority == 'high' else "🟡" if priority == 'medium' else "🟢"
report += f"{i}. {emoji} **{rec['title']}**
"
report += f" {rec['description']}
"
return report
数据分析报告
import pandas as pd
import numpy as np
from datetime import datetime
def generate_data_analysis_report(df, title="数据分析报告"):
"""
生成全面的数据分析报告
"""
report = f"""# {title}
**日期:** {datetime.now().strftime('%Y-%m-%d')}
**数据集:** {len(df):,} 行 × {len(df.columns)} 列
---
## 目录
1. [数据集概览](#数据集概览)
2. [数据质量](#数据质量)
3. [统计摘要](#统计摘要)
4. [分布](#分布)
5. [相关性](#相关性)
6. [见解](#见解)
---
## 数据集概览
### 基本信息
- **总行数:** {len(df):,}
- **总列数:** {len(df.columns)}
- **内存使用:** {df.memory_usage(deep=True).sum() / 1024**2:.2f} MB
- **重复行:** {df.duplicated().sum():,}
### 列信息
| 列 | 类型 | 非空 | 唯一 | 样本值 |
|--------|------|----------|--------|---------------|
"""
for col in df.columns:
dtype = str(df[col].dtype)
non_null = df[col].count()
unique = df[col].nunique()
samples = df[col].dropna().head(3).tolist()
sample_str = ", ".join(str(s) for s in samples)
report += f"| {col} | {dtype} | {non_null:,} | {unique:,} | {sample_str} |
"
report += """
---
## 数据质量
### 缺失值
"""
missing = df.isnull().sum()
if missing.sum() > 0:
report += "| 列 | 缺失计数 | 缺失 % |
"
report += "|--------|---------------|----------|
"
for col in missing[missing > 0].index:
count = missing[col]
pct = (count / len(df)) * 100
report += f"| {col} | {count:,} | {pct:.1f}% |
"
else:
report += "✅ 未检测到缺失值。
"
report += "
### 数据类型问题
"
# 检查潜在类型问题
type_issues = []
for col in df.select_dtypes(include=['object']):
# 检查列是否为数值型
try:
pd.to_numeric(df[col], errors='raise')
type_issues.append(f"- `{col}` 似乎是数值型但存储为字符串")
except:
pass
# 检查列是否为日期时间型
try:
pd.to_datetime(df[col], errors='raise')
if df[col].str.contains(r'\d{4}-\d{2}-\d{2}').any():
type_issues.append(f"- `{col}` 似乎是日期时间型但存储为字符串")
except:
pass
if type_issues:
report += "
".join(type_issues) + "
"
else:
report += "✅ 未检测到数据类型问题。
"
report += """
---
## 统计摘要
### 数值列
"""
# 添加数值列的统计
numeric_cols = df.select_dtypes(include=[np.number]).columns
if len(numeric_cols) > 0:
stats = df[numeric_cols].describe()
report += stats.to_markdown() + "
"
# 添加额外统计
report += "
### 额外统计
"
report += "| 列 | 中位数 | 众数 | 标准差 | 方差 |
"
report += "|--------|--------|------|---------|----------|
"
for col in numeric_cols:
median = df[col].median()
mode = df[col].mode().iloc[0] if not df[col].mode().empty else "N/A"
std = df[col].std()
var = df[col].var()
report += f"| {col} | {median:.2f} | {mode} | {std:.2f} | {var:.2f} |
"
report += """
### 分类列
"""
categorical_cols = df.select_dtypes(include=['object']).columns
if len(categorical_cols) > 0:
for col in categorical_cols[:5]: # 限制为前5个
report += f"
#### {col}
"
value_counts = df[col].value_counts().head(10)
report += "| 值 | 计数 | 百分比 |
"
report += "|-------|-------|------------|
"
for value, count in value_counts.items():
pct = (count / len(df)) * 100
report += f"| {value} | {count:,} | {pct:.1f}% |
"
report += """
---
## 分布
"""
# 分析数值列的分布
for col in numeric_cols[:5]: # 限制为前5个
report += f"
### {col} 分布
"
q1 = df[col].quantile(0.25)
q2 = df[col].quantile(0.50)
q3 = df[col].quantile(0.75)
iqr = q3 - q1
# 检测异常值
lower_bound = q1 - 1.5 * iqr
upper_bound = q3 + 1.5 * iqr
outliers = df[(df[col] < lower_bound) | (df[col] > upper_bound)]
report += f"""
**四分位数:**
- Q1 (25%): {q1:.2f}
- Q2 (50%, 中位数): {q2:.2f}
- Q3 (75%): {q3:.2f}
- IQR: {iqr:.2f}
**异常值:** {len(outliers)} ({len(outliers)/len(df)*100:.1f}%)
- 下界: {lower_bound:.2f}
- 上界: {upper_bound:.2f}
"""
report += """
---
## 相关性
"""
if len(numeric_cols) > 1:
corr_matrix = df[numeric_cols].corr()
report += "
### 相关矩阵
"
report += corr_matrix.to_markdown() + "
"
# 查找强相关性
report += "
### 强相关性 (|r| > 0.7)
"
strong_corr = []
for i in range(len(corr_matrix.columns)):
for j in range(i+1, len(corr_matrix.columns)):
corr_val = corr_matrix.iloc[i, j]
if abs(corr_val) > 0.7:
col1 = corr_matrix.columns[i]
col2 = corr_matrix.columns[j]
strong_corr.append((col1, col2, corr_val))
if strong_corr:
for col1, col2, corr_val in strong_corr:
direction = "正相关" if corr_val > 0 else "负相关"
report += f"- **{col1}** ↔ **{col2}**: {corr_val:.3f} ({direction})
"
else:
report += "未找到强相关性。
"
report += """
---
## 见解
"""
# 生成见解
insights = generate_insights(df)
for insight in insights:
report += f"### {insight['title']}
"
report += f"{insight['description']}
"
if 'details' in insight:
for detail in insight['details']:
report += f"- {detail}
"
report += "
"
return report
def generate_insights(df):
"""生成数据见解"""
insights = []
# 见解：完整性
missing_pct = (df.isnull().sum().sum() / (len(df) * len(df.columns))) * 100
if missing_pct < 1:
status = "优秀"
emoji = "✅"
elif missing_pct < 5:
status = "良好"
emoji = "👍"
else:
status = "需注意"
emoji = "⚠️"
insights.append({
"title": f"{emoji} 数据完整性: {status}",
"description": f"总体数据完整性为 {100-missing_pct:.1f}%，缺失值占 {missing_pct:.1f}%。",
"details": [
f"总单元格数: {len(df) * len(df.columns):,}",
f"缺失单元格数: {df.isnull().sum().sum():,}"
]
})
# 见解：重复记录
dup_count = df.duplicated().sum()
if dup_count > 0:
insights.append({
"title": f"⚠️ 发现重复记录",
"description": f"发现 {dup_count:,} 重复行 ({dup_count/len(df)*100:.1f}% 数据集)",
"details": [
"考虑移除重复以提高分析准确性",
"审查业务逻辑以处理重复"
]
})
return insights
性能报告
def generate_performance_report(metrics, baseline=None):
"""
生成性能比较报告
"""
report = f"""# 性能报告
**生成时间:** {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}
---
## 摘要
"""
if baseline:
report += "### 与基线比较
"
report += "| 指标 | 当前 | 基线 | 变化 | 状态 |
"
report += "|--------|---------|----------|--------|--------|
"
for metric_name, current_value in metrics.items():
if metric_name in baseline:
baseline_value = baseline[metric_name]
change = ((current_value - baseline_value) / baseline_value) * 100
if abs(change) < 5:
status = "🟢 稳定"
elif change > 0:
status = "🟢 改进" if is_improvement(metric_name, change) else "🔴 退化"
else:
status = "🔴 退化" if is_improvement(metric_name, change) else "🟢 改进"
report += f"| {metric_name} | {current_value:.2f} | {baseline_value:.2f} | {change:+.1f}% | {status} |
"
else:
report += "### 当前指标
"
report += "| 指标 | 值 | 状态 |
"
report += "|--------|-------|--------|
"
for metric_name, value in metrics.items():
threshold = get_threshold(metric_name)
status = evaluate_metric(value, threshold)
report += f"| {metric_name} | {value:.2f} | {status} |
"
report += """
---
## 详细分析
"""
for metric_name, value in metrics.items():
report += f"### {metric_name}
"
if baseline and metric_name in baseline:
baseline_value = baseline[metric_name]
change = ((value - baseline_value) / baseline_value) * 100
report += f"- **当前:** {value:.2f}
"
report += f"- **基线:** {baseline_value:.2f}
"
report += f"- **变化:** {change:+.1f}%
"
if abs(change) > 10:
report += f"⚠️ 检测到显著变化。 "
report += "审查可能影响此指标的近期变化。
"
else:
report += f"- **值:** {value:.2f}
"
return report
def is_improvement(metric_name, change):
"""根据指标类型确定变化是否为改进"""
# 这些指标越低越好
lower_is_better = ['response_time', 'error_rate', 'latency', 'load_time']
for pattern in lower_is_better:
if pattern in metric_name.lower():
return change < 0
return change > 0
HTML报告生成
def generate_html_report(data, title="报告", template="default"):
"""
生成带样式的HTML报告
"""
# CSS样式
css = """
<style>
body {
font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Oxygen, Ubuntu, Cantarell, sans-serif;
line-height: 1.6;
color: #333;
max-width: 1200px;
margin: 0 auto;
padding: 20px;
background: #f5f5f5;
}
.report-container {
background: white;
padding: 40px;
border-radius: 8px;
box-shadow: 0 2px 4px rgba(0,0,0,0.1);
}
h1 {
color: #2c3e50;
border-bottom: 3px solid #3498db;
padding-bottom: 10px;
}
h2 {
color: #34495e;
margin-top: 30px;
border-left: 4px solid #3498db;
padding-left: 10px;
}
h3 {
color: #7f8c8d;
}
table {
width: 100%;
border-collapse: collapse;
margin: 20px 0;
}
th {
background: #3498db;
color: white;
padding: 12px;
text-align: left;
font-weight: 600;
}
td {
padding: 10px 12px;
border-bottom: 1px solid #ecf0f1;
}
tr:hover {
background: #f8f9fa;
}
.metric-card {
background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
color: white;
padding: 20px;
border-radius: 8px;
margin: 10px 0;
box-shadow: 0 4px 6px rgba(0,0,0,0.1);
}
.metric-value {
font-size: 2em;
font-weight: bold;
}
.metric-label {
font-size: 0.9em;
opacity: 0.9;
}
.badge {
display: inline-block;
padding: 4px 12px;
border-radius: 12px;
font-size: 0.85em;
font-weight: 600;
}
.badge-success {
background: #2ecc71;
color: white;
}
.badge-warning {
background: #f39c12;
color: white;
}
.badge-danger {
background: #e74c3c;
color: white;
}
.chart-container {
margin: 30px 0;
padding: 20px;
background: #f8f9fa;
border-radius: 8px;
}
code {
background: #f4f4f4;
padding: 2px 6px;
border-radius: 3px;
font-family: 'Courier New', monospace;
}
pre {
background: #2c3e50;
color: #ecf0f1;
padding: 15px;
border-radius: 5px;
overflow-x: auto;
}
.timestamp {
color: #7f8c8d;
font-size: 0.9em;
}
</style>
"""
# 生成HTML内容
html = f"""
<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>{title}</title>
{css}
<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
</head>
<body>
<div class="report-container">
<h1>{title}</h1>
<p class="timestamp">生成时间: {datetime.now().strftime('%B %d, %Y at %I:%M %p')}</p>
{generate_html_content(data)}
</div>
</body>
</html>
"""
return html
def generate_html_content(data):
"""从数据生成HTML内容"""
html = ""
# 关键指标部分
if 'metrics' in data:
html += "<h2>关键指标</h2>"
html += '<div style="display: grid; grid-template-columns: repeat(auto-fit, minmax(250px, 1fr)); gap: 15px;">'
for metric in data['metrics']:
html += f"""
<div class="metric-card">
<div class="metric-label">{metric['name']}</div>
<div class="metric-value">{metric['value']}</div>
</div>
"""
html += "</div>"
# 表格数据
if 'table' in data:
html += "<h2>数据表格</h2>"
html += generate_html_table(data['table'])
# 图表
if 'charts' in data:
for chart in data['charts']:
html += f'<h2>{chart["title"]}</h2>'
html += '<div class="chart-container">'
html += generate_chart_html(chart)
html += '</div>'
return html
def generate_html_table(table_data):
"""从数据生成HTML表格"""
html = "<table>"
# 表头
if 'headers' in table_data:
html += "<thead><tr>"
for header in table_data['headers']:
html += f"<th>{header}</th>"
html += "</tr></thead>"
# 行
html += "<tbody>"
for row in table_data.get('rows', []):
html += "<tr>"
for cell in row:
html += f"<td>{cell}</td>"
html += "</tr>"
html += "</tbody>"
html += "</table>"
return html
def generate_chart_html(chart_data):
"""生成Chart.js图表"""
chart_id = f"chart_{abs(hash(chart_data['title']))}"
html = f'<canvas id="{chart_id}" width="400" height="200"></canvas>'
html += f"""
<script>
var ctx = document.getElementById('{chart_id}').getContext('2d');
var chart = new Chart(ctx, {{
type: '{chart_data.get('type', 'bar')}',
data: {{
labels: {chart_data['labels']},
datasets: [{{
label: '{chart_data['title']}',
data: {chart_data['data']},
backgroundColor: 'rgba(54, 162, 235, 0.5)',
borderColor: 'rgba(54, 162, 235, 1)',
borderWidth: 2
}}]
}},
options: {{
responsive: true,
maintainAspectRatio: true,
scales: {{
y: {{
beginAtZero: true
}}
}}
}}
}});
</script>
"""
return html
Markdown表格
def generate_markdown_table(data, headers=None, alignment=None):
"""
从数据生成markdown表格
alignment: 'left', 'center', 'right' 的列表
"""
if not data:
return ""
# 如果未提供，自动检测表头
if headers is None:
if isinstance(data[0], dict):
headers = list(data[0].keys())
else:
headers = [f"列 {i+1}" for i in range(len(data[0]))]
# 生成表头行
table = "| " + " | ".join(str(h) for h in headers) + " |
"
# 生成对齐行
if alignment is None:
alignment = ['left'] * len(headers)
align_chars = {
'left': ':--',
'center': ':-:',
'right': '--:'
}
table += "| " + " | ".join(align_chars.get(a, ':--') for a in alignment) + " |
"
# 生成数据行
for row in data:
if isinstance(row, dict):
row_data = [str(row.get(h, '')) for h in headers]
else:
row_data = [str(cell) for cell in row]
table += "| " + " | ".join(row_data) + " |
"
return table
# 使用示例
data = [
{"name": "John", "age": 30, "city": "New York"},
{"name": "Jane", "age": 25, "city": "San Francisco"},
{"name": "Bob", "age": 35, "city": "Chicago"}
]
table = generate_markdown_table(
data,
headers=['姓名', '年龄', '城市'],
alignment=['left', 'right', 'left']
)
图表和可视化
import matplotlib.pyplot as plt
import seaborn as sns
def generate_chart(data, chart_type='bar', title='图表', output='chart.png'):
"""
从数据生成图表
"""
plt.figure(figsize=(10, 6))
plt.style.use('seaborn-v0_8-darkgrid')
if chart_type == 'bar':
plt.bar(data['labels'], data['values'])
elif chart_type == 'line':
plt.plot(data['labels'], data['values'], marker='o', linewidth=2)
elif chart_type == 'pie':
plt.pie(data['values'], labels=data['labels'], autopct='%1.1f%%')
elif chart_type == 'scatter':
plt.scatter(data['x'], data['y'], alpha=0.6)
plt.title(title, fontsize=16, fontweight='bold')
plt.tight_layout()
plt.savefig(output, dpi=300, bbox_inches='tight')
plt.close()
return output
# 用于markdown报告
def embed_chart_in_markdown(chart_path, alt_text="图表"):
"""生成markdown图像嵌入"""
return f"
"
PDF导出
from markdown import markdown
from weasyprint import HTML
def markdown_to_pdf(markdown_text, output_path='report.pdf', css=None):
"""
将markdown转换为PDF
"""
# 将markdown转换为HTML
html_content = markdown(markdown_text, extensions=['tables', 'fenced_code'])
# 包装在HTML文档中
html_doc = f"""
<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8">
<style>
{css if css else get_default_pdf_css()}
</style>
</head>
<body>
{html_content}
</body>
</html>
"""
# 转换为PDF
HTML(string=html_doc).write_pdf(output_path)
def get_default_pdf_css():
"""PDF导出的默认CSS"""
return """
body {
font-family: Arial, sans-serif;
line-height: 1.6;
color: #333;
max-width: 800px;
margin: 0 auto;
padding: 20px;
}
h1 {
color: #2c3e50;
border-bottom: 2px solid #3498db;
padding-bottom: 10px;
}
table {
width: 100%;
border-collapse: collapse;
margin: 20px 0;
}
th, td {
border: 1px solid #ddd;
padding: 8px;
text-align: left;
}
th {
background-color: #3498db;
color: white;
}
code {
background: #f4f4f4;
padding: 2px 4px;
border-radius: 3px;
}
pre {
background: #f4f4f4;
padding: 10px;
border-radius: 5px;
overflow-x: auto;
}
"""
报告模板
TEMPLATES = {
'executive': {
'sections': ['summary', 'key_metrics', 'recommendations'],
'style': 'concise'
},
'technical': {
'sections': ['overview', 'detailed_analysis', 'code_examples', 'metrics'],
'style': 'comprehensive'
},
'comparison': {
'sections': ['baseline', 'current', 'differences', 'trends'],
'style': 'comparative'
}
}
def generate_from_template(data, template_name='executive'):
"""从模板生成报告"""
template = TEMPLATES.get(template_name, TEMPLATES['executive'])
report = f"# {template_name.title()} 报告
"
for section in template['sections']:
report += generate_section(data, section, template['style'])
return report
最佳实践
- 清晰结构化报告，带有目录
- 使用视觉层次（标题、表格、图表）
- 包含时间戳，所有报告
- 为长报告添加执行摘要
- 使用一致格式化 贯穿始终
- 包括数据源 和方法论
- 添加可操作建议
- 使用图表表示趋势，表格表示详细数据
- 导出到多种格式（MD、HTML、PDF）
- 尽可能自动化报告生成
注释
- 保持报告专注和可操作
- 为数据类型使用适当可视化
- 包括摘要和详细视图
- 版本控制报告模板
- 测试不同数据大小的PDF导出
- 考虑HTML报告的可访问性
- 为HTML报告使用响应式设计
- 为性能缓存生成的图表