skills/vercel-labs/native/zero-native/Gen Agent Trust Hub

zero-native

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • Dynamic Instruction Loading: The skill is designed to fetch its core logic and reference materials at runtime using the zero-native skills get command. This pattern allows the agent to access up-to-date documentation and implementation details directly from the tool.
  • External Package Reference: The instructions reference the zero-native and skills packages for installation via NPM or execution via NPX. This is a standard procedure for installing development tools.
  • Tool Output Processing: Content retrieved from the CLI is ingested into the agent context to provide orientation and implementation tasks. While this facilitates dynamic updates, it creates an entry point where tool output influences agent behavior.
  • Command Execution Scope: To function as intended, the skill utilizes Bash to execute zero-native, npx, and zig commands, which are used for project setup, building, and orientation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 12:11 AM
Security Audit — agent-trust-hub — zero-native