agenthook-ai-influencer-ugc-video
Pass
Audited by Gen Agent Trust Hub on Jul 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the
getagenthookpackage from the NPM registry to enable the requiredagenthookCLI functionality. - [COMMAND_EXECUTION]: The skill relies on executing shell commands via the
agenthookCLI for core tasks such as authentication, balance checks, and media generation. - [PROMPT_INJECTION]: The skill processes user-supplied prompts and external media URLs through third-party AI models via the AgentHook API, which represents an indirect prompt injection surface.
- [CREDENTIALS_UNSAFE]: While the skill manages sensitive API keys, it includes robust safety instructions that prohibit the agent from displaying, logging, or echoing these keys. It prioritizes a browser-based 'device-login' flow to keep secrets out of the conversation transcript.
Audit Metadata