The Agent Skills Directory

[COMMAND_EXECUTION]: The skill automatically executes shell commands to install software dependencies via pip and brew, and to perform image processing using the macOS sips utility.\n- [REMOTE_CODE_EXECUTION]: The skill dynamically generates and executes Python scripts at runtime to produce formatted PDF documents from extracted text.\n- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through malicious text embedded in document images. Ingestion points: Text extracted via OCR from user-provided images as described in SKILL.md. Boundary markers: Absent; the skill is instructed to proceed automatically without requesting confirmation or reviewing content. Capability inventory: Shell command execution, dynamic script generation/execution, and file system access. Sanitization: Absent; the skill does not mention validation or sanitization of OCR output before processing.\n- [DATA_EXFILTRATION]: High-impact data handling. The skill is designed to process highly sensitive identity and financial documents (passports, bank statements), increasing the severity of any potential injection or compromise despite no explicit exfiltration logic.

visa-doc-translate