retro
Warn
Audited by Socket on Aug 21, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the core retrospective behavior is plausible, but the skill's footprint is broader than needed because it depends on unverifiable local gstack binaries, persistent analytics, and telemetry handling outside the repo-analysis task. No confirmed credential theft or overtly malicious payload is visible, but install/execution trust and data-flow transparency are not proportionate enough to call benign.
Confidence: 81%Severity: 78%
Audit Metadata