pza-settings
Warn
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [DATA_EXFILTRATION]: The skill accesses a configuration file in the user's home directory (~/.claude/pza-settings.json) using dynamic context injection. This automatically reads the file into the agent context upon skill load.
- [COMMAND_EXECUTION]: The skill uses dynamic context injection placeholders to run shell commands like
whichandcatwhen the file is first accessed. - [COMMAND_EXECUTION]: The skill instructs the agent to update settings by interpolating user-provided arguments into a
node -eshell command. This pattern is susceptible to command injection if the agent does not properly sanitize the input before executing the resulting shell command.
Audit Metadata