csharp-patterns

Pass

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides informational content and code snippets for C# 13/15 and .NET 10/11 features. All external links point to official Microsoft documentation, and the code examples for secret management use secure patterns like configuration providers and managed identities instead of hardcoded credentials.
  • [PROMPT_INJECTION]: The .NET AI integration section contains code examples demonstrating prompt interpolation. While this represents a potential attack surface for indirect prompt injection if implemented with untrusted data in a production environment, its inclusion here is educational and limited to documenting the use of specific AI libraries.
  • Ingestion points: The SummarizeAsync method and Semantic Kernel prompt template in SKILL.md.
  • Boundary markers: Absent in the simplified educational snippets.
  • Capability inventory: None. The skill is purely documentation and contains no executable scripts or tools.
  • Sanitization: Absent in the simplified educational snippets.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 4, 2026, 10:03 AM