ai-ugc-ad-video
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes 'Script helpers' that direct the agent to fetch and analyze text from untrusted external sources like Reddit, YouTube, and Amazon to generate marketing insights.
- Ingestion points: The agent is instructed to ingest data from community platforms, including reviews and comments (found in
references/prompts.md). - Boundary markers: While the skill uses bracketed placeholders for interpolation, it lacks explicit delimiters or instructions to ignore embedded commands within the external data being processed.
- Capability inventory: The skill environment includes access to shell tools (
curl,ffmpeg,ffprobe) and file system access (Read,Write,Edit) through its allowed tools. - Sanitization: No explicit sanitization or filtering logic is provided for the text retrieved from external web sources before it is processed by the agent.
Audit Metadata