paw-cra-creative-director

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes a script (scripts/tool-discovery.py) that uses subprocess.run to verify the availability and versions of CLI tools such as ffmpeg and agent-browser. This is a legitimate environment check performed with safe argument passing.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions and references for well-known creative APIs and services, including fal.ai, Pexels, and ElevenLabs. It documents how the user can configure these via local configuration files according to platform standards.
  • [DATA_EXFILTRATION]: The skill manages a local memory system within the .pawbytes/ directory to store brand profiles, briefs, and campaign status. It facilitates the creation of a social-auth.json file via agent-browser for authenticated research, but this is presented as a user-controlled feature for automation and does not involve unauthorized data transmission.
  • [SAFE]: No prompt injection, obfuscation, or persistence mechanisms were detected. The skill's behavior is consistent with its stated purpose of creative orchestration.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 05:14 PM