openspec-onboard

SUSPICIOUS. The core OpenSpec onboarding flow is broadly consistent with its stated purpose and uses an official CLI, but the skill expands trust by invoking other skills and suggesting installation of another skill, while also processing untrusted repo content and enabling write/implementation actions. No clear credential theft or explicit exfiltration is present, so this is not malicious, but the transitive-skill and prompt-injection exposure make it medium risk.