NYC

vercel-observability

Pass

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: LOWNO_CODE
Full Analysis
  • [NO_CODE] (SAFE): The skill consists exclusively of markdown documentation and metadata. No code (Python, JavaScript, shell scripts) or configuration files are present in the provided files.
  • [EXTERNAL_DOWNLOADS] (SAFE): All references are static URLs to official Vercel documentation. No automated downloads, remote code execution, or package installations (npm/pip) are defined.
  • [PROMPT_INJECTION] (SAFE): The instructions are purely descriptive and do not contain any attempts to bypass agent constraints, reveal system prompts, or redirect behavior.
  • [DATA_EXFILTRATION] (SAFE): There are no mechanisms for file access, credential retrieval, or network communication within the skill.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 15, 2026, 10:43 PM