yoga-with-adriene

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill provides operational boundaries for the agent but does not attempt to bypass safety guidelines or override fundamental system behavior.
  • [DATA_EXFILTRATION]: No sensitive file access or network-based exfiltration patterns were identified. The mention of an external domain (ansaur.com) is for informational purposes for the user.
  • [COMMAND_EXECUTION]: No shell commands, subprocess calls, or system-level operations are present in the skill content.
  • [EXTERNAL_DOWNLOADS]: The skill does not perform any automated downloads or fetch remote scripts for execution.
  • [REMOTE_CODE_EXECUTION]: There are no mechanisms for dynamic code loading or remote execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 10:56 PM
Security Audit — agent-trust-hub — yoga-with-adriene