Answer: NVIDIA described “NVIDIA-verified agent skills” as a way to add transparency and a chain-of-trust to reusable agent skills used in workflows, with skills that are cataloged, scanned, signed, and documented with a machine-readable skill card. NVIDIA said the verification process includes a scanning step using SkillSpector to check software risks and agent-specific risks such as hidden instructions, prompt injection, trigger abuse, tool poisoning, and mismatches between declared purpose and behavior.
NVIDIA said the approach builds on the agentskills.io open skills specification and mentioned compatibility goals across tools including Claude Code, Codex, and Cursor. NVIDIA also referenced OWASP guidance and MITRE ATLAS in describing governance grounding for the scanning and validation signals.
Source: NVIDIA Developer