antigravity-manager
Warn
Audited by Socket on Sep 14, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill's behavior is broadly coherent with its stated purpose, but that purpose is itself high-risk: it centralizes Google/Anthropic web session tokens, stores local API/admin secrets, and turns unofficial web sessions into proxy API access. Same-org distribution reduces malware confidence, yet the raw-script installers, mutable Windows install path, and explicit handling/recovery of stored secrets make the overall skill unsafe to trust casually.
Confidence: 85%Severity: 78%
Audit Metadata