ugc-ad
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill features a data ingestion surface where user-provided media is analyzed and interpolated into scripts for batch runners. 1. Ingestion points: User-provided media reviewed via '$media-analysis' in SKILL.md Review Mode. 2. Boundary markers: The skill uses production locks and specific shot table structures to delimit input. 3. Capability inventory: Uses '$image-batch-runner' and '$video-batch-runner' for asset generation. 4. Sanitization: Relies on 'Claim Boundaries' constraints and human approval gates. This surface is assessed as safe due to the highly structured templates.
- [SAFE]: The skill consists of instructional Markdown files with no executable code, remote dependencies, or hidden scripts.
- [SAFE]: No evidence of prompt injection, obfuscation, or persistence mechanisms was found; the instructions incorporate safety-oriented constraints against generating misleading content.
- [SAFE]: Data exfiltration and unauthorized tool usage are not present; the skill respects the defined tools and workflow.
Audit Metadata