The Agent Skills Directory

[COMMAND_EXECUTION]: The skill generates and executes Python scripts using PIL and reportlab to assemble PDFs. It also uses the system-level sips utility for converting HEIC images to PNG.
[EXTERNAL_DOWNLOADS]: Instructions include the installation of various common libraries such as easyocr, pytesseract, reportlab, and pillow. These are retrieved from standard registries but represent external dependencies required for the skill's functionality.
[INDIRECT_PROMPT_INJECTION]: The skill processes text extracted via OCR from user-provided documents. There is a risk that malicious instructions hidden within a document (e.g., a fake bank statement) could be interpreted by the AI during the translation or script generation phase.
Ingestion points: OCR-extracted text from image files provided by the user (as described in SKILL.md).
Boundary markers: None. The instructions do not provide delimiters or context isolation for the extracted text.
Capability inventory: Subprocess execution for sips, package installation via pip, and dynamic execution of agent-generated Python scripts.
Sanitization: None. The skill does not specify any validation or filtering of the text extracted from documents before it is processed by the agent.

visa-doc-translate