The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes standard .NET SDK commands such as dotnet format and dotnet --info. These are legitimate tools for the intended purpose of formatting code and verifying environment setup.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it reads and processes external files from the repository being analyzed.
Ingestion points: AGENTS.md, solution/project files, and .editorconfig files are read from the repository context.
Boundary markers: No specific delimiters or safety instructions are defined to wrap the ingested content.
Capability inventory: The skill is capable of executing shell commands via the dotnet CLI.
Sanitization: No specific sanitization or filtering logic is present for the data read from project files. Despite this surface, the skill provides specific guidance to the agent to 'Prefer the SDK-provided dotnet format command instead of inventing custom format scripts,' which acts as a procedural safeguard against malicious instructions embedded in the project files.

format