deep-research
Warn
Audited by Socket on May 20, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill’s stated purpose matches research orchestration, but its footprint is broad: autonomous background subagents, browser automation, Bash execution, and arbitrary web-content ingestion. The biggest issue is indirect prompt-injection risk from combining untrusted external content with tools that can execute commands and write files; secondary concern is the unpinned npx chromux fallback. No clear credential harvesting or exfiltration behavior is shown, so this is not confirmed malware.
Confidence: 84%Severity: 74%
Audit Metadata