The Agent Skills Directory

[DYNAMIC_CONTEXT_INJECTION]: The skill utilizes the !command syntax to execute a find command at load time to automatically identify .NET solution files (.sln, .slnx) in the current directory. This is a legitimate use of the feature to provide immediate environmental context relevant to the skill's purpose.
[INDIRECT_PROMPT_INJECTION]: The skill involves reading and parsing file contents using tools like grep and the dotnet CLI to identify entry points, configuration, and dependencies. This creates a surface where malicious instructions embedded in project files could influence the agent's behavior.
Ingestion points: Files like Program.cs, .csproj, .sln, and appsettings.json (SKILL.md).
Boundary markers: None identified; instructions do not explicitly warn the agent to ignore instructions embedded in the project files.
Capability inventory: The skill uses find, grep, ls, and the dotnet CLI to read files and query project metadata.
Sanitization: None identified; the skill directly processes the output of these commands.
[COMMAND_EXECUTION]: The skill provides numerous shell command examples using find, grep, and dotnet for solution exploration. These are standard developer operations and do not include patterns for exfiltration or sensitive credential access.

dotnet-solution-navigation