The Agent Skills Directory

[COMMAND_EXECUTION]: The skill guides the agent to use standard dotnet CLI commands, such as dotnet new install, dotnet new list, and dotnet build, to install and verify the custom templates being authored. These are legitimate operations within a local development environment.
[EXTERNAL_DOWNLOADS]: The instructions include references to official Microsoft documentation and the .NET templating engine wiki on GitHub for schema and configuration guidance. These are trusted, well-known resources for .NET developers.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and bootstrap content from external project files (.csproj) and configuration files (template.json). While this creates a potential surface for indirect prompt injection, it is the fundamental purpose of the skill and no specific vulnerabilities or malicious redirection were observed.
Ingestion points: Analysis of local .csproj and template.json files provided by the user.
Boundary markers: Not explicitly specified; the skill relies on standard project structure parsing.
Capability inventory: Shell command execution via dotnet CLI and local file system modifications for template generation.
Sanitization: No explicit sanitization logic is provided in the instructions, as it targets trusted local project sources.
[SAFE]: No malicious patterns, obfuscation, persistence mechanisms, or unauthorized privilege escalation attempts were found. The skill aligns with documented .NET development best practices.

template-authoring