efficient-frontier

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFENO_CODEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill consists of instructional content for managing AI model orchestration. No malicious patterns, obfuscation, or unauthorized data access mechanisms were detected.
  • [NO_CODE]: The skill is composed of Markdown files providing guidelines and does not include any executable scripts or binary payloads.
  • [EXTERNAL_DOWNLOADS]: The documentation references the installation of the @agent-native/skills package via npx, which is the standard deployment mechanism for this ecosystem.
  • [PROMPT_INJECTION]: The provided instructions define a workflow for delegating tasks between models but do not include directives to bypass safety filters or ignore system constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 08:15 AM
Security Audit — agent-trust-hub — efficient-frontier