cartoon-ad
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves ingesting untrusted content from users which could potentially contain instructions designed to override agent behavior.
- Ingestion points: The skill processes user-supplied brand briefs, product concepts, and existing advertisements provided for auditing purposes (SKILL.md).
- Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore instructions that might be embedded within the ingested briefs or analyzed media.
- Capability inventory: Based on the ingested content, the skill triggers actions across multiple tools, including
$media-analysis,$image-batch-runner, and$video-batch-runner(SKILL.md, references/workflow-handoff.md). - Sanitization: While the instructions emphasize stylistic and legal boundaries (such as the Non-Imitation Boundary), there is no mention of technical sanitization or validation of the input data before it is interpolated into production scripts.
Audit Metadata