skillui

SUSPICIOUS. The capability broadly matches the stated design-extraction purpose, and there is no clear credential harvesting or exfiltration flow in the skill text. However, the skill's core behavior requires executing a third-party CLI from a personal publisher and then trusting generated `SKILL.md`/`CLAUDE.md` content inside Claude, which creates meaningful supply-chain and transitive-trust risk disproportionate to a simple documentation skill. No confirmed malware indicators are present.