youtube-production
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection because it ingests and propagates untrusted user-provided data and files through a multi-agent system.
- Ingestion points: User-provided topics, keywords, and existing files (scripts or briefs) are saved to the workspace directory and shared among the strategist, scriptwriter, designer, and SEO agents.
- Boundary markers: The orchestration instructions lack requirements for delimiters or specific instructions to the sub-agents to ignore instructions embedded within the user data.
- Capability inventory: The agents possess the ability to write to the local file system and communicate via inter-agent messaging.
- Sanitization: The workflow does not specify sanitization or validation steps for external content before it is interpolated into the agent prompts.
Audit Metadata