The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script (scripts/generate_emergency_card.py) to produce HTML versions of medical cards. The source code of this script was not provided in the analyzed file set, rendering its implementation and safety unverifiable.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted medical record data from multiple files without adequate security boundaries, posing a vulnerability if input data contains malicious instructions. * Ingestion points: Sensitive medical record files including data/profile.json, data/allergies.json, data/medications/medications.json, and various surgical or discharge summaries found via globbing. * Boundary markers: Absent; health data is extracted and interpolated directly into card templates without delimiters or protective instructions. * Capability inventory: File system operations (readFile, saveFile, glob) and shell command execution (python). * Sanitization: Absent; the agent does not validate or sanitize the text content of ingested health records before processing.
[EXTERNAL_DOWNLOADS]: The HTML output format incorporates external assets from well-known content delivery networks (Tailwind CSS and Lucide Icons). These are reputable services but represent external dependencies for the final generated card.

emergency-card