The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions and scripts reference the installation of standard, widely-recognized Python libraries including mplfinance, matplotlib, plotly, pandas, numpy, and scipy. These are official packages from public registries used for the skill's primary purpose of data analysis and visualization.
[COMMAND_EXECUTION]: The documentation includes shell commands for environment setup, specifically using the uv package manager to install dependencies. These commands are transparent and consistent with standard developer workflows.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data in the form of Pandas DataFrames and Series to generate charts. While this constitutes a data ingestion surface, the risk is minimized as the skill focuses on rendering graphical outputs (PNG, SVG, HTML) rather than parsing the data to influence agent control logic. No unsafe interpolation or capability chains were detected.
[DATA_EXPOSURE_AND_EXFILTRATION]: The scripts include functionality to save generated charts to the local file system. This is a core feature of the skill, and no evidence was found of sensitive file access or network transmission to unauthorized domains.

trading-visualization