open-claudecode-research
Warn
Audited by Socket on Jul 30, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The stated research purpose only partly matches the footprint, but the critical issue is trust: it asks users to clone and run an unofficial reconstructed Claude Code runtime from a personal repo, then supply Anthropic credentials and optionally route them through arbitrary proxies. That combination is disproportionate and creates strong supply-chain and credential-interception risk even without explicit malicious code shown.
Confidence: 91%Severity: 90%
Audit Metadata