azure-ai-vision-imageanalysis-java
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill enables the agent to process external content via OCR (VisualFeatures.READ), which introduces a potential surface for indirect prompt injection.
- Ingestion points: The skill ingests data from local files via BinaryData.fromFile() and from remote URLs via the analyzeFromUrl() method.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to isolate extracted text or to disregard any commands that might be embedded within the images.
- Capability inventory: The skill utilizes file system access to read images and performs network operations to interact with Azure AI services.
- Sanitization: There is no evidence of sanitization, filtering, or validation of the text returned by the OCR process before it is added to the agent's context.
Audit Metadata