The Agent Skills Directory

[COMMAND_EXECUTION]: The batch_generate.py script executes the companion generate_qr.py script using subprocess.check_call. The implementation correctly uses a list-based argument structure rather than a shell string, which is a secure practice that prevents shell injection.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from CSV files for its batch generation feature. It incorporates the following security controls:
Ingestion points: Data including IDs, URLs, and labels are read from a CSV file in batch_generate.py.
Boundary markers: The workflow instructions include a step to validate URLs before generation.
Capability inventory: The skill possesses file-writing capabilities (PathWrite and PIL.save) and local command execution capabilities (subprocess.check_call).
Sanitization: The validate_url function ensures destination links use http/https and contain a domain, while EscapeXML sanitizes caption labels to prevent injection in SVG output files.
[SAFE]: The project dependencies (qrcode, pillow) are standard, reputable packages for image and QR code generation. The skill does not attempt to access sensitive system directories, perform network exfiltration, or maintain persistence.

qr-code-generator