The Agent Skills Directory

[SAFE]: The skill is primarily instructional documentation for the Microsoft Semantic Kernel framework. It contains no executable malicious code, obfuscation, or unauthorized network operations.
[CREDENTIALS_UNSAFE]: The skill demonstrates secure credential management by using Environment.GetEnvironmentVariable and builder.Configuration instead of hardcoding API keys.
[EXTERNAL_DOWNLOADS]: All external URLs point to official Microsoft documentation and official GitHub repositories, which are recognized as trusted sources.
[DATA_EXFILTRATION]: No evidence of sensitive data access or unauthorized exfiltration. Network operations are limited to standard AI service endpoints and local development services (localhost) for vector databases.
[PROMPT_INJECTION]: The skill utilizes Handlebars templates and structured YAML prompt configurations, which are standard patterns for managing LLM inputs and reducing the risk of accidental prompt manipulation.

dotnet-semantic-kernel