The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data inputs such as DataFrames, CSV files, and Newick phylogenetic trees (Tree.nwk) as described in chart-recipes.md and itol-reference.md.
Ingestion points: Data ingestion occurs through variable interpolation (e.g., df) and file system reads of user-supplied data files.
Boundary markers: No explicit boundary markers or instructions to ignore embedded prompts within the data are present.
Capability inventory: The skill is configured with Bash, Read, Grep, and Glob tools, and its scripts execute file-write operations (plt.savefig).
Sanitization: There is no evidence of sanitization or escaping of data-derived strings (e.g., species labels) before they are rendered in plots or used in shell searches.
[EXTERNAL_DOWNLOADS]: The documentation references official scientific resources including matplotlib.org, seaborn.pydata.org, itol.embl.de, and colorbrewer2.org. These are well-known, trusted domains for academic and scientific visualization tools.
[COMMAND_EXECUTION]: The skill uses shell utilities like grep (to extract species names from tree files) and identify (to verify image dimensions), which are standard operations for the skill's visualization and export tasks.

data-visualization