The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides code examples in references/excel-charts.md that utilize subprocess.run to execute a local helper script named recalc.py. This script is intended to facilitate the recalculation of Excel formulas after data modification, which is a standard procedure for the openpyxl library when results are not cached.
[PROMPT_INJECTION]: The skill is designed to ingest and visualize external datasets, which creates a potential surface for indirect prompt injection (Category 8). Since it processes untrusted data (CSV, Excel, etc.) into visual artifacts, developers should ensure that the agent context is protected by boundary markers when interpreting the resulting data summaries.
[EXTERNAL_DOWNLOADS]: The skill correctly identifies and utilizes external dependencies from well-known and reputable services. It fetches the D3.js library from the jsDelivr CDN and geographic GeoJSON data from the official Plotly GitHub repository. These sources are considered trusted and appropriate for the skill's domain.

data-viz