The Agent Skills Directory

[SAFE]: The skill is strictly advisory, providing expert guidance for JetBrains IDE JVM tuning without executing any automated system changes or external scripts.
[DATA_EXPOSURE_EXFILTRATION]: The skill instructs the agent to ask the user for diagnostic output from standard JVM tools (e.g., jcmd, thread dumps, GC logs). While these sources can contain sensitive information like environment variables or code snippets, the skill includes explicit warnings to the user about these risks (e.g., in memory-options.md and performance-validation.md). No automated exfiltration patterns were identified.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided diagnostic data. Although this represents an ingestion surface for potentially malicious content embedded in logs or command output, the risk is mitigated by the skill's focused diagnostic workflow and its role in generating human-readable recommendations rather than performing automated execution.
[EXTERNAL_DOWNLOADS]: All external references target official documentation and source repositories from trusted organizations, specifically JetBrains and Oracle (OpenJDK), for the purpose of verifying JVM flag behaviors.

jetbrains-vmoptions