可视化时间序列 · Time Series Visualizer

它解决什么问题 · The problem

时间序列图一多，读者只看到线，看不到结论。这个技能把「看图」变成「看结论」：先拆趋势、周期、异常，再给每张图配一句人话结论。

With too many time-series charts, readers see lines but no conclusions. This skill turns chart-reading into conclusion-reading: it decomposes trend, cycle, and anomaly first, then attaches a plain-language takeaway to every chart.

这个技能做什么 · What it does

它先确认数据口径与时间粒度，再把序列拆成趋势、周期与异常三类，输出结论先行的分析：每张图配结论说明、跨平台绘图脚本（Python/R）与异常点核查清单。

It confirms data definitions and time granularity first, then decomposes each series into trend, cycle, and anomaly and outputs a conclusion-first analysis: every chart comes with a takeaway, cross-platform plotting scripts (Python/R), and an anomaly checklist.

为什么比裸提示词强 · Why it beats raw prompts

• 结论先行，先看懂再细看 / Conclusions first, so readers get it before zooming in
• 每张图配可运行脚本，直接复现 / Every chart ships with runnable scripts for instant reproduction
• 自动标出异常点，附核查建议 / Auto-flags anomalies with verification suggestions
• 跨平台脚本说明环境差异 / Scripts note platform differences between environments

典型用例 · Typical use cases

• 业务指标日报、周报的可视化 / Visualizing business metrics for daily and weekly reports
• 服务器监控数据的趋势与异常分析 / Trend and anomaly analysis of server monitoring data
• 销售、流量等时间序列的周期性拆解 / Decomposing sales and traffic seasonality
• 给管理层汇报用的结论加图材料 / Executive-ready charts with conclusions attached

局限 · Limitations

提供分析与脚本，不连接你的数据源自动取数；数据质量（缺失、口径不一）需先在源端处理。

It delivers analysis and scripts but does not connect to your data sources; missing values and inconsistent definitions must be handled at the source.