skills/jimliu/science-skills/boltz/Gen Agent Trust Hub

boltz

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill utilizes the 'boltz' and 'cuequivariance_ops_torch' Python packages, which are standard components for this model. These are fetched from standard package registries.
  • [DATA_EXFILTRATION]: The skill documents the use of the '--use_msa_server' flag, which sends protein sequence data to the ColabFold API (api.colabfold.com) to generate Multiple Sequence Alignments. This is a common and established practice in structural biology workflows and is transparently disclosed in the skill metadata and documentation.
  • [COMMAND_EXECUTION]: Executes the 'boltz' command-line tool to perform structure predictions based on user-provided YAML configuration files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 01:21 AM
Security Audit — agent-trust-hub — boltz