首页/AI 智能体外部交互/firecrawl-dashboard-reporting
F

firecrawl-dashboard-reporting

by @firecrawlv
4.6(120)

此技能利用 Firecrawl 浏览器从分析仪表盘和内部网络工具中自动提取关键指标。它能处理认证访问、设置日期范围、抓取 KPI 卡片、表格及标签,并支持跨平台指标汇总和结构化数据报告,显著提升数据获取效率和准确性。

dashboard-reportingweb-scrapingdata-extractionautomationmetrics-analysisGitHub
安装方式
npx skills add https://github.com/firecrawl/firecrawl-workflows --skill firecrawl-dashboard-reporting
compare_arrows

Before / After 效果对比

1 组
使用前

手动访问多个仪表盘,耗时且易出错,难以实现高频更新。

使用后

AI 自动抓取并汇总数据,报告生成迅速,确保数据实时性和准确性。

SKILL.md

Firecrawl Dashboard Reporting

Use this to extract visible metrics from dashboards the user can legitimately access.

Onboarding Interview

Infer dashboard URLs, metrics, date range, and output format from context. If dashboard targets are clear and accessible, proceed immediately.

Ask at most 1-3 concise questions only if blocked, such as the dashboard URLs, auth/profile requirement, or date range.

Firecrawl Collection Plan

Use Firecrawl browser for authenticated dashboards and UI interaction:

  • open each dashboard
  • set or verify date range
  • extract visible KPI cards, tables, and labels
  • click tabs, expand sections, and scroll tables
  • use export/download buttons only when appropriate and allowed

If login has expired, ask the user to re-authenticate rather than attempting to bypass access controls.

Parallel Work

If appropriate, use sub-agents or equivalent parallel task runners. Split by dashboard platform or metric category. Each researcher should return metrics, units, period, source URL, and caveats.

Final Deliverable

# Dashboard Report

## Summary
[Highlights, alerts, trends]

## Metrics By Dashboard
[Platform, metric, value, unit, change, period]

## Tables Or Exports
[Captured tables/files and what they contain]

## Notes And Caveats
[Auth issues, chart-only data, unavailable metrics]

## Rerun Inputs
workflow: firecrawl-dashboard-reporting
dashboards: [urls]
date_range: [range]
metrics: [list]
output: [json/markdown]

JSON Shape

Use reportedAt, dateRange, dashboards[], metrics[], tables[], exports[], and summary.

Quality Bar

  • Extract actual numbers, not just chart labels.
  • Note when a chart cannot be read precisely.
  • Preserve date ranges and source URLs.

用户评价 (0)

发表评价

效果
易用性
文档
兼容性

暂无评价

统计数据

安装量31.6K
评分4.6 / 5.0
版本
更新日期2026年10月23日
对比案例1 组

用户评分

4.6(120)
5
37%
4
43%
3
13%
2
5%
1
2%

为此 Skill 评分

0.0

兼容平台

🤖claude-code

时间线

创建2026年5月22日
最后更新2026年10月23日
🎁 Agent 知识卡片
调研问卷