- name
SKILL.md
技能檔案保留在發布者的儲存庫中。
在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add DougTrajano/pydantic-ai-skills --skill data-analysis -a cursor -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add DougTrajano/pydantic-ai-skills --skill data-analysis -a claude-code -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add DougTrajano/pydantic-ai-skills --skill data-analysis -a codex -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add DougTrajano/pydantic-ai-skills --skill data-analysis -a grok -y在 Grok Bot 聊天中傳送複製的訊息。Bot 會從 GitHub 取得這個 SKILL.md,並加入你所有 Bot 共用的技能中。
Install this skill: https://github.com/DougTrajano/pydantic-ai-skills/tree/HEAD/examples/skills/data-analysis來自發布者
Answers questions about a bundled sales dataset (resources/sales.csv): 96 rows
covering six months, four regions, four product categories and three sales
channels.
This is the reference skill for exercising sandbox executors. Both scripts are standard library only and read their data from the skill folder, so they do real work with no network, no third-party packages and no host access — exactly the shape a sandbox is meant to run.
aggregateaggregateprofile_datasetaggregate with filtersReports dataset shape, column names, summary statistics for every numeric column (count, sum, mean, median, stdev, min, max) and distinct-value counts for the categorical ones.
column (optional): Profile a single numeric column instead of all of themExits 2 for an unknown column.
Groups rows and aggregates a numeric column, optionally filtered and truncated.
group-by (required): Column to group by — month, region, category, channelmetric (optional): Numeric column to aggregate, default revenueagg (optional): sum (default), mean, median, min, max, countwhere (optional, repeatable): Filter as column=value, e.g. region=northtop (optional): Keep only the highest N groupsExits 2 for an unknown column or malformed filter, 1 when no rows match.
Revenue by region, highest first:
Top 3 categories by units sold, online only:
Average revenue per month in the west:
resources/sales.csv is generated sample data, included so the skill is
self-contained and deterministic. It is not real sales data.
Both scripts produce byte-identical output on the host and under
LocalSandboxScriptExecutor or OpenSandboxScriptExecutor, including the
../resources/sales.csv read through the skill root and non-zero exit codes for
bad arguments. That equivalence is the point: swapping the executor changes where
the script runs, not what the model sees.
Profile and aggregate a bundled sales dataset - group revenue by region, category, channel or month, apply filters, and compute sums, means, medians and ranges. Use for questions about sales figures,
---name: data-analysisdescription: Profile and aggregate a bundled sales dataset - group revenue by region, category, channel or month, apply…---# examples/skills/data-analysis/SKILL.md總覽
Profile and aggregate a bundled sales dataset - group revenue by region, category, channel or month, apply filters, and compute sums, means, medians and ranges. Use for questions about sales figures, trends, or which segments perform best.
常見問題
從發布者下載 examples/skills/data-analysis/SKILL.md,或複製路徑後開啟儲存庫。PlugMCP 只收錄這項技能,不代管檔案。
發布者將它存放在 https://github.com/DougTrajano/pydantic-ai-skills。
Claude、Cursor、Codex 和 Grok Build 可以讀取你從發布者儲存庫加入的 SKILL.md。在聊天中傳送技能的 GitHub 資料夾連結,Grok Bot 就會安裝它。