The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill takes 'C# code or intent' as input to generate and execute scripts. This creates an attack surface where an attacker providing data to the agent could embed instructions that manipulate the generated script's logic.
Ingestion points: 'Inputs' section in SKILL.md (C# code or intent).
Boundary markers: Absent; there are no instructions to delimit or ignore instructions within the provided intent.
Capability inventory: Shell execution via dotnet CLI, file system writes, and permission modification via chmod.
Sanitization: Absent; the skill does not specify any validation or sanitization for the generated code.
[COMMAND_EXECUTION]: The skill relies on executing shell commands to run scripts, check versions, and clean up build artifacts.
Evidence: Use of dotnet --version, dotnet <file>.cs, dotnet clean, and dotnet project convert.
[PRIVILEGE_ESCALATION]: The instructions include modifying file permissions to make generated scripts executable.
Evidence: Recommendation to use chmod +x hello.cs for Unix shebang support.
[DYNAMIC_EXECUTION]: The core functionality of the skill involves the dynamic generation of C# source files which are then compiled and executed at runtime.
Evidence: Step-by-step workflow for writing a .cs file and running it via the dotnet runtime.
[EXTERNAL_DOWNLOADS]: The skill references external documentation from a well-known service.
Evidence: Provides a link to official Microsoft documentation at https://learn.microsoft.com/en-us/dotnet/core/sdk/file-based-apps.

csharp-scripts