The Agent Skills Directory

[SAFE]: The skill is a technical reference for building data visualizations. No malicious patterns such as obfuscation, exfiltration, or persistence were found.
[INDIRECT_PROMPT_INJECTION]: The skill architecture handles external data ingestion for visualization, creating a surface for potential prompt injection. * Ingestion points: Data is ingested via skillLibraryData.ts and example datasets in assets/example-datasets/. * Boundary markers: Semantic HTML and ARIA labels provide structure, though no security-specific markers are present. * Capability inventory: File reading/writing via scripts and client-side CSV downloads are supported. * Sanitization: Relies on established libraries (Recharts/D3) and accessibility validation.

visualizing-data