NVIDIA introduced “NVIDIA-verified agent skills” and described them as a way to add transparency and provenance to the capability layer of autonomous agents.
NVIDIA said verified skills are cataloged, scanned, signed, and documented with a “skill card.” NVIDIA said the approach builds on an open skills specification so a SKILL.md can work across tools including Claude Code, Codex, and Cursor.
NVIDIA said its SkillSpector scanning checks conventional software risks and agent-specific risks such as hidden instructions, prompt injection, trigger abuse, excessive agency, and tool poisoning.
Source: NVIDIA Developer