employee-surveillance-dpia
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill does not contain any instructions intended to override AI safety constraints or manipulate agent behavior.
- [DATA_EXFILTRATION]: No network connectivity or access to sensitive local file paths (such as credentials or SSH keys) was found.
- [REMOTE_CODE_EXECUTION]: There is no evidence of remote script fetching, package installation, or dynamic execution of untrusted code.
- [COMMAND_EXECUTION]: The provided scripts do not invoke shell commands, system processes, or administrative privileges.
- [CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or other sensitive secrets were detected in the instructions or scripts.
- [SAFE]: The skill is entirely self-contained and operates within the bounds of standard data processing for privacy assessments.
Audit Metadata