deeplake
Warn
Audited by Socket on Mar 28, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: The skill’s storage and query capabilities match its purpose, and data flows primarily target official DeepLake endpoints. The main issue is install/execution trust: it asks the agent to pipe a same-brand but not publicly verified CLI installer into bash, then use that CLI for authenticated mount/sync operations. This is not clearly malicious, but the unverifiable binary plus credential use and persistent filesystem scope make the skill high-risk.
Confidence: 82%Severity: 82%
Audit Metadata