vision-framework
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to extract information from external images and video feeds, which inherently creates an indirect prompt injection attack surface.
- Ingestion points: Untrusted data enters the processing pipeline via
cgImage,pixelBuffer, andsampleBufferparameters in various functions acrossSKILL.mdandreferences/vision-requests.md. - Boundary markers: The skill does not implement delimiters or specific instructions to the agent to ignore potentially malicious content found within the scanned text or barcodes.
- Capability inventory: The skill provides data extraction patterns only; it does not include dangerous capabilities like arbitrary command execution or external network exfiltration based on the scanned data.
- Sanitization: No input sanitization or validation logic is provided for the extracted data strings.
Audit Metadata