The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the sips system utility to convert HEIC images to PNG. It also dynamically generates Python code using the PIL and reportlab libraries to create the final PDF document, which is then executed.
[EXTERNAL_DOWNLOADS]: The instructions direct the agent to install several third-party libraries and tools from official registries using pip and brew, including easyocr, pytesseract, reportlab, and pyobjc-frameworks. These are well-known dependencies for OCR and PDF generation.
[PROMPT_INJECTION]: The skill processes untrusted data extracted via OCR from user-provided images, which creates an indirect prompt injection surface. 1. Ingestion points: Text extracted from image files processed in SKILL.md. 2. Boundary markers: No delimiters or safety instructions are used to separate the extracted text from the agent's internal instructions. 3. Capability inventory: The skill has the ability to execute system commands (sips) and run dynamically generated Python scripts (SKILL.md). 4. Sanitization: No sanitization or validation is applied to the extracted text before it is used for translation.

visa-doc-translate