csharp-refactoring
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection Surface]: The skill interacts with external codebase files and executes build commands, which could potentially ingest malicious content if the repository is untrusted.
- Ingestion points: The agent reads C# source code, project files, and repository metadata from the user's workspace.
- Boundary markers: The instructions direct the agent to use binding-aware tools and Roslyn providers, maintaining structural boundaries during the refactoring process.
- Capability inventory: The skill is authorized to modify files, execute shell commands (
dotnet build,dotnet test), and interact with LSP plugins. - Sanitization: Validation is managed through compiler checks and test suites to ensure that refactoring does not introduce unintended behaviors.
- [External Resource Reference]: The skill references an LSP configuration file from a well-known service to support semantic code analysis. This is an informational reference to standard tool settings.
Audit Metadata