csharp-dotnet

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from the local environment, including C# source code (.cs, .razor), project definitions (.csproj, .props, .targets), and MSBuild binary logs (.binlog).
  • Ingestion points: The skill instructions command the agent to read and analyze source code and build artifacts in the review, fix-build, and diagnose-runtime-issue workflows.
  • Boundary markers: The instructions lack specific guidance for the agent to ignore or delimit potentially malicious instructions embedded in code comments, string literals, or build metadata.
  • Capability inventory: The skill utilizes powerful CLI tools including dotnet build, dotnet restore, dotnet msbuild, and dotnet tool install, and is instructed to modify project files directly.
  • Sanitization: No explicit sanitization or validation of the input files (code, logs) is described in the workflows.
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install official .NET diagnostic tools (dotnet-counters, dotnet-trace, dotnet-dump) globally. These are well-known tools provided by Microsoft and distributed via the official NuGet registry.
  • [COMMAND_EXECUTION]: The skill makes extensive use of the .NET CLI and MSBuild for project analysis and diagnostics. These commands (e.g., dotnet msbuild -pp, dotnet restore, dotnet build -bl) are standard for the skill's stated purpose of assisting with .NET development and troubleshooting.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 03:14 PM
Security Audit — agent-trust-hub — csharp-dotnet