The Agent Skills Directory

[SAFE]: No security issues detected. The skill contains educational material, templates, and a benign Python script.
[DATA_EXPOSURE]: The included script phi_scope_checker.py operates locally on user-provided text files. It does not perform network operations, exfiltrate data, or access sensitive system directories.
[COMMAND_EXECUTION]: The skill instructions and workflows use a local Python script for keyword scanning. No unauthorized privilege escalation or dangerous shell command patterns were found.
[INDIRECT_PROMPT_INJECTION]: While the phi_scope_checker.py script processes untrusted user input (product descriptions), it uses static regular expressions for scanning and does not execute the input as code or pass it to an LLM prompt, effectively eliminating the risk of injection-based attacks.

healthtech-advisor