Skills
Navigate
skills/immamdouhaboammar/marketingskills/ad-creative/Gen Agent Trust Hub

ad-creative

Fail

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill directs users to download external tools and repositories from unverified sources, including the installation of gooseworks via npx and cloning the voicebox repository from GitHub.
  • [REMOTE_CODE_EXECUTION]: The skill facilitates the execution of remote code by providing instructions to install and run third-party software from sources not listed as trusted vendors, such as the gooseworks package and the voicebox repository.
  • [DYNAMIC_EXECUTION]: The skill uses the Remotion framework for React-based video rendering and generates HTML review pages that include JavaScript for dynamic data rendering.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted customer reviews and ad comments from external files to influence its generation logic, creating a surface for indirect prompt injection.
  • [COMMAND_EXECUTION]: The skill relies on various CLI tools for ad platform integrations, including a local script located at tools/clis/google-ads.js.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 19, 2026, 08:11 PM