amazon-tiktok-signal
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted text content from external TikTok video data via the
asin_video_searchtool. Maliciously crafted content within video titles or descriptions could attempt to influence the agent's logic or bias the final business recommendations. - Ingestion points: Video metadata (titles, descriptions, creator information) processed in
SKILL.mdand evaluated using criteria inreferences/statistics-and-judgment.md. - Boundary markers: Absent. The instructions do not define delimiters or provide specific warnings to the agent to ignore instructions embedded within the retrieved data fields.
- Capability inventory: Limited impact. The skill is focused on analysis and reporting; it does not utilize high-risk tools such as file writing, shell execution, or arbitrary network requests within its provided instructions.
- Sanitization: Absent. There is no evidence of filtering or validation to strip potentially adversarial instructions from the data before it is processed by the model.
Audit Metadata