ralph-loops
Warn
Audited by Socket on Mar 1, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
Overall, the skill appears consistent with its stated purpose of guiding an autonomous loop agent, with one notable risk: the kill $PPID instruction could disrupt parent processes or automation pipelines if executed inappropriately. Absent hidden or credentialed capabilities, the material is largely benign in intent but warrants caution in deployment to ensure the exit command cannot be triggered in unintended contexts. Recommend restricting the exit mechanism to controlled environments and documenting the exact process boundaries to avoid accidental termination of critical tooling.
Confidence: 75%Severity: 75%
Audit Metadata