The Agent Skills Directory

[SAFE]: The skill serves as a legitimate technical guide for .NET developers, providing standard implementation patterns for HTTP endpoints.
[PROMPT_INJECTION]: No attempts to override system instructions or bypass safety filters were detected. The instructional language is consistent with technical documentation.
[DATA_EXFILTRATION]: No code snippets demonstrate accessing sensitive files (e.g., .env, .aws/credentials) or exfiltrating data to external domains. All endpoint examples use local routing patterns.
[REMOTE_CODE_EXECUTION]: There are no patterns involving the download or execution of remote scripts via curl, wget, or similar tools.
[COMMAND_EXECUTION]: The skill does not contain any shell command execution or subprocess spawning.
[OBFUSCATION]: Analysis of the text and code snippets revealed no Base64 encoding, zero-width characters, or hidden URLs designed to evade detection.
[DYNAMIC_EXECUTION]: While the skill uses Activator.CreateInstance for endpoint discovery, this is a standard and safe reflection pattern in .NET development when scoped to types within the local assembly (typeof(Program).Assembly).

minimal-api