ghidra-mcp
Warn
Audited by Socket on Oct 13, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill is internally coherent for reverse engineering, but it equips the agent with high-risk offensive security capabilities and relies on unverifiable local absolute-path tooling, especially the GhidraMCP bridge. I found no clear credential theft or external exfiltration, so this is better classified as high-risk/vulnerable rather than confirmed malware.
Confidence: 86%Severity: 78%
Audit Metadata