proteinmpnn

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists exclusively of high-level instructional text for a scientific workflow. No executable code, scripts, or automated tool configurations are present.
  • [PROMPT_INJECTION]: The instructions are domain-specific and do not attempt to override agent safety protocols or bypass system constraints.
  • [DATA_EXFILTRATION]: There are no commands to access sensitive system files, environment variables, or credentials. No network operations are defined.
  • [REMOTE_CODE_EXECUTION]: The skill does not download external packages or execute remote scripts. The mention of verifying execution routes is a conceptual workflow step rather than a functional command.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 04:04 AM
Security Audit — agent-trust-hub — proteinmpnn