The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided content through the ${selection} variable.
Ingestion points: The ${selection} placeholder in SKILL.md is used to intake code for analysis.
Boundary markers: None present to delimit user code from instructions.
Capability inventory: The skill explicitly instructs the agent to only provide a review and not make changes. No external tools, file-system operations, or network capabilities are requested or utilized.
Sanitization: No sanitization of the input selection is performed.
[SAFE]: The skill follows best practices by explicitly restricting its scope to code review only ("Do not make any changes to the code, just provide a review"). No malicious patterns such as obfuscation, credential theft, or unauthorized command execution were found.

dotnet-design-pattern-review