self-review-apply
Warn
Audited by Socket on Feb 25, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
Benign. The fragment describes a local, structured self-review application workflow for parsing XML feedback, optionally loading image attachments for context, and applying mechanical code changes. There is no evident credential handling, external network communication, or data exfiltration. The multi-agent orchestration concept (subagents per file) is within a typical automation pattern and does not by itself imply harm. Overall security risk is low, with no malware indicators detected based on the provided content.
Confidence: 75%Severity: 75%
Audit Metadata