The Agent Skills Directory

[SAFE]: The skill is composed as a documentation-only instruction set for performing isolation reviews. It does not contain any functional code, scripts, or tool invocations that could pose a security risk.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest system architecture and configuration data for analysis. However, it includes explicit constraints stating that no user-provided values are used in commands or file paths, and it performs no active operations, effectively mitigating risks associated with processing potentially malicious configuration data.

isolation-review