The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill includes instructions and examples for processing external data (e.g., pandas DataFrames) to generate plots. While this represents a data ingestion surface, it is the primary intended function of the tool and is handled using standard scientific libraries without dangerous execution vectors.
Ingestion points: The skill takes user-provided datasets as input for plotting functions in SKILL.md and references/matplotlib_examples.md.
Boundary markers: No specific delimiters or boundary instructions are included for untrusted data.
Capability inventory: The skill is capable of writing files to the local file system (PDF, PNG, etc.) via matplotlib.figure.Figure.savefig, as shown in scripts/figure_export.py.
Sanitization: The skill relies on standard library plotting logic and does not implement additional string sanitization for input data.
[DYNAMIC_EXECUTION]: The script scripts/style_presets.py contains a function create_style_template that writes a matplotlib style configuration file (.mplstyle) to the disk. This is a standard configuration task for visualization tools and does not involve executing arbitrary generated code.
[METADATA_POISONING]: A discrepancy exists between the author listed in the YAML frontmatter ('K-Dense Inc.') and the expected author context ('eturkes'). This appears to be a neutral branding choice rather than a deceptive or malicious pattern.

scientific-visualization