agent-tools
Warn
Audited by Socket on Apr 14, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the core inference.sh CLI usage appears aligned with the stated platform and uses same-org infrastructure, so this is not confirmed malware. However, the skill has medium-high security risk because it relies on an unpinned remote installer, uploads local files to cloud services, enables autonomous public/social actions, and instructs transitive installation of additional third-party-managed skills.
Confidence: 87%Severity: 76%
Audit Metadata