amz-ai-image-policy-compliance
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFENO_CODEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides a qualitative framework for auditing images and does not involve any automated execution or system interaction.
- [NO_CODE]: No scripts or binaries are included; the skill is composed purely of Markdown instructions and templates.
- [EXTERNAL_DOWNLOADS]: The skill references a WhatsApp invite link for a community group. WhatsApp is a well-known service, and the link is provided for informational purposes without automated retrieval or execution.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided image data, which presents a surface for indirect prompt injection if an image contains malicious text. However, the risk is negligible as the skill lacks any capabilities (such as tools, network access, or file system modifications) that could be exploited by injected instructions.
- Ingestion points: User-provided image set (SKILL.md)
- Boundary markers: Absent
- Capability inventory: None (zero tools)
- Sanitization: Absent
Audit Metadata