The Agent Skills Directory

[Data Exposure & Exfiltration] (SAFE): The skill reads and writes to a specific local file .agent/.user-profile.json. There are no network calls or attempts to access sensitive system directories or credentials.
[Remote Code Execution] (SAFE): The skill does not employ dynamic execution functions like eval() or exec(). All logic is contained within static Python scripts using standard libraries.
[Indirect Prompt Injection] (SAFE): The skill processes user input via the preference_extractor.py and stores results in a JSON profile. It does not use this data to influence its own execution logic, though other skills reading this profile should treat the data as untrusted.

nav-profile