headless-ghidra-frida-verify

SUSPICIOUS: the skill’s reverse-engineering capabilities broadly match its purpose, but it grants an AI agent high-impact Frida-based execution/attachment abilities and requires an unverifiable external wrapper CLI (`ghidra-agent-cli`). There is no clear evidence of credential theft or external exfiltration, so this is not confirmed malware, but it is a high-risk skill due to offensive tooling and install-trust concerns.