The Agent Skills Directory

Dynamic Code Execution: The skill is designed to compile and execute C# code directly from source files using the dotnet CLI. This is the primary function of the skill, providing a powerful capability to run arbitrary logic on the host system.
Command Line Operations: The workflow utilizes various system-level commands including dotnet --version, dotnet run, dotnet clean, and chmod +x. These commands are used to manage the execution environment and file permissions for the generated scripts.
Indirect Prompt Injection Surface: The skill ingests "C# code or intent" which is then transformed into executable files. This processing of untrusted input into executable code creates an attack surface where embedded instructions in the input could potentially execute unintended logic.
Ingestion points: The skill accepts "C# code or intent" as a primary input (SKILL.md).
Boundary markers: There are no explicit boundary markers or instructions to the agent to ignore potentially malicious embedded content within the provided C# code.
Capability inventory: The skill possesses the ability to execute dotnet commands, write .cs files to the filesystem, and modify file execution permissions (SKILL.md).
Sanitization: No specific sanitization, validation, or escaping mechanisms for the C# code input are documented in the skill's workflow.
External Dependency Management: Through the #:package directive, the skill can download and include NuGet packages. While this is standard for .NET development, it involves fetching and integrating external code from remote package registries.

csharp-scripts