lifestyle-ad
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and process untrusted external data which could influence the behavior of downstream production runners.
- Ingestion points: Workflow Step 1 in SKILL.md specifies that the agent should read user-supplied brief components, including brand documents, product claims, personas, and reference scripts.
- Boundary markers: The instructions do not define clear delimiters or use 'ignore embedded instructions' warnings when processing these external data sources.
- Capability inventory: The skill possesses the capability to trigger production actions via the $video-batch-runner, $image-batch-runner, and $workflow-creation tools, which generate media assets or reusable platform workflows.
- Sanitization: The instructions across the skill and reference files do not specify any validation, filtering, or sanitization of the input content before it is interpolated into the creative context for the production handoff.
Audit Metadata