ad-creative
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from web search results and user-provided performance metrics, creating a potential surface for indirect injection.
- Ingestion points: Web search queries for market research and user-uploaded performance data (SKILL.md).
- Boundary markers: Absent. No specific delimiters or safety instructions are defined for processing external data.
- Capability inventory: File system writes and shell command execution for tool integrations (SKILL.md).
- Sanitization: Absent. There are no instructions for sanitizing or validating ingested data.
- [COMMAND_EXECUTION]: The skill provides examples for running local CLI tools and third-party rendering engines like Remotion.
- [EXTERNAL_DOWNLOADS]: The documentation references and provides examples for using various well-known AI services and APIs, including Google Gemini, ElevenLabs, and OpenAI.
Audit Metadata