connecting-to-sbx-sandboxes

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists entirely of documentation regarding the sbx CLI tool and SSH configuration. It does not contain executable scripts or automated commands beyond descriptive examples.
  • [SAFE]: References to sensitive file paths such as ~/.ssh/config and environment variables like ANTHROPIC_API_KEY are provided in a purely educational context to explain how the sbx tool manages security boundaries and isolation.
  • [SAFE]: The documentation explicitly warns users about potential security risks, such as credential injection in specific applications (Claude Desktop) and the implications of opting into environment variable forwarding.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 04:43 PM
Security Audit — agent-trust-hub — connecting-to-sbx-sandboxes