dotnet-audit-trail

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements an AuditLogInterceptor that records entity changes into an AuditLog table. This creates a surface for indirect prompt injection, where malicious strings stored in application entities are propagated into audit logs that may be processed by agents or administrators later.
  • Ingestion points: Data is ingested from entity properties in AuditLogInterceptor.cs during the EF Core SavingChangesAsync event.
  • Boundary markers: The skill uses JsonSerializer.Serialize to wrap changed values, providing a structural JSON boundary.
  • Capability inventory: The skill uses Entity Framework Core to perform write operations to the database.
  • Sanitization: While values are serialized to JSON, the skill does not perform content-based filtering or sanitization of the values themselves.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 12:04 PM
Security Audit — agent-trust-hub — dotnet-audit-trail