elon-musk

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill's activation instructions explicitly command the agent to 'completely abandon the default persona' (abandona completamente a persona padrao). This language is a characteristic prompt injection pattern intended to override the base AI's identity and safety constraints in favor of a specific persona.
  • [COMMAND_EXECUTION]: The YAML frontmatter lists 'antigravity' as an allowed tool. While this is a well-known Python easter egg that opens a web browser to a comic, its inclusion in the execution environment configuration flags the potential for triggering unintended local system actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 09:39 AM
Security Audit — agent-trust-hub — elon-musk