sinch-functions-dotnet

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill identifies that data originating from external webhooks is untrusted and poses an injection risk. It provides explicit instructions to prevent exploitation.
  • Ingestion points: The skill processes user-controlled data via request.Menu?.Input (SKILL.md), data.Cli (references/context-services.md), and MessageInboundEvent (references/conversation-webhooks.md).
  • Boundary markers: The instructions explicitly state: "Validate before use; never interpolate into prompts, shell commands, or SQL."
  • Capability inventory: The skill has access to Context.Database (SQLite) and Context.Storage (File system/S3).
  • Sanitization: The skill demonstrates and mandates the use of parameters with Microsoft.Data.Sqlite and Dapper to mitigate SQL injection risks.
  • [SAFE]: The skill references official vendor packages (Sinch.Functions.Runtime) and established, well-known libraries (Dapper, Microsoft.Data.Sqlite, Microsoft.AspNetCore.Mvc).
  • [SAFE]: All external references and documentation links are directed to the trusted vendor domain developers.sinch.com.
  • [SAFE]: The skill encourages secure secret management by instructing users to use dotnet user-secrets or the sinch secrets CLI tool instead of hardcoding credentials.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 04:29 PM
Security Audit — agent-trust-hub — sinch-functions-dotnet