matrixscan-count-net-ios

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an inherent attack surface for indirect prompt injection (Category 8) because it is designed to ingest and interpret untrusted user-provided content, such as application source code and project configuration files.
  • Ingestion points: The skill processes user-provided C# view controllers (EmptyViewController.cs, IntegratedViewController.cs), project files (.csproj), and metadata files (Info.plist).
  • Boundary markers: The skill uses standard markdown code block delimiters to separate user data from its internal instruction context.
  • Capability inventory: The skill is authorized to suggest code modifications, generate new implementation logic, and provide dependency management advice based on the ingested content.
  • Sanitization: There is no evidence of explicit sanitization or pre-processing filters to detect or neutralize malicious instructions that might be embedded within the user-provided code files.
  • [EXTERNAL_DOWNLOADS]: The skill implements a process to fetch current package metadata from the official NuGet registry to ensure accurate dependency management.
  • Evidence: The integration and migration guides (references/integration.md, references/migration.md) instruct the agent to use the WebFetch tool to retrieve the latest stable version of the Scandit.DataCapture.Barcode package from https://www.nuget.org/packages/Scandit.DataCapture.Barcode/ or the NuGet V3 API.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 03:17 AM
Security Audit — agent-trust-hub — matrixscan-count-net-ios