id-capture-net-ios
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a legitimate developer integration guide for the Scandit ID Capture SDK and does not contain malicious code or instructions.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from identity documents (passports, IDs), which is its primary function.
- Ingestion points: Data is ingested via the IIdCaptureListener and stored in CapturedId objects.
- Boundary markers: The skill relies on the SDK's internal parsing and typed result model.
- Capability inventory: The skill is restricted to camera operations and UI alerts; it lacks capabilities for arbitrary file access or network exfiltration.
- Sanitization: Raw data is converted into structured fields (e.g., DateResult), providing a level of validation.
- [CREDENTIALS_SAFE]: The skill provides explicit instructions to prevent the exposure of license keys, including masking keys in chat and using secure MCP-based retrieval instead of filesystem searches.
Audit Metadata