The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill's documentation and installation scripts (install.sh, install.ps1) utilize piped shell execution commands (curl | bash and irm | iex) to fetch and execute installers from the author's GitHub repository.
[EXTERNAL_DOWNLOADS]: The installer downloads prebuilt binaries and the skill package from the vendor's repository (CodeAlive-AI/ai-driven-development) during setup.
[COMMAND_EXECUTION]: The agent is instructed to execute a CLI tool (bin/csharp-refactor) or invoke the .NET SDK via dotnet run to perform refactorings on the project.
[PROMPT_INJECTION]: Indirect Prompt Injection Surface:
Ingestion points: The skill ingests untrusted data by reading C# source code files from the user's repository via the Roslyn API.
Boundary markers: The skill does not implement boundary markers or instructions for the agent to ignore potentially malicious instructions embedded in comments or strings within the code files.
Capability inventory: The agent has the capability to modify the filesystem by running the refactoring tool which writes changes to disk.
Sanitization: There is no evidence of content sanitization or validation of the source code content before it is processed and returned to the agent context.

refactoring-csharp