solublempnn

Warn

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill clones a software repository and downloads model weights from a third-party GitHub source (https://github.com/dauparas/ProteinMPNN.git) that is not part of the established trusted organization list.
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to execute shell commands to install dependencies using pip, clone the external repository, and invoke a Python script with various arguments.
  • [REMOTE_CODE_EXECUTION]: The skill implements a 'download-and-run' pattern where it fetches an external codebase and executes a runner script (protein_mpnn_run.py) immediately. This behavior introduces a security risk if the remote repository were to be compromised.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 14, 2026, 10:23 PM
Security Audit — agent-trust-hub — solublempnn