The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill follows standard development practices for .NET environments and integrates official tooling.
[EXTERNAL_DOWNLOADS]: Fetches the JetBrains.ReSharper.GlobalTools package from the official NuGet registry using the dotnet tool command. This is a recognized development utility from a well-known vendor and follows repository-local or global installation standards.
[COMMAND_EXECUTION]: Executes standard .NET SDK and JetBrains CLI commands (dotnet build, jb inspectcode, jb cleanupcode) locally within the repository context. These operations are essential for the skill's functionality and are performed without elevated privileges.
[DATA_EXFILTRATION]: No evidence of unauthorized data transfer or access to sensitive credentials (such as SSH keys, AWS configs, or .env secrets) was found. The skill operates exclusively on project-specific source files and configuration.
[PROMPT_INJECTION]: The skill processes untrusted content from the repository as part of its analysis. Ingestion points: .cs source files, .sln/.csproj project files, .DotSettings configuration files, and AGENTS.md instructions. Boundary markers: No specific delimiters or ignore-instructions were identified. Capability inventory: Includes dotnet tool installation, solution building, and in-place code cleanup. Sanitization: No explicit sanitization of ingested content was observed. While this represents a surface for indirect prompt injection, it is within acceptable parameters for code-editing tools.

resharper-clt