alterlab-rfdiffusion

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a wrapper for RFdiffusion, a well-known protein design tool developed by the RosettaCommons academic organization. No malicious behavior, obfuscation, or unauthorized data access patterns were identified.
  • [SAFE]: All external references, such as the RosettaCommons/RFdiffusion GitHub repository and internal AlterLab suite tools (e.g., alterlab-remote-compute, alterlab-proteinmpnn), are legitimate resources consistent with the skill's stated academic purpose.
  • [SAFE]: The skill follows security best practices by using restricted tool scoping in the allowed-tools frontmatter, limiting shell execution to specific python and uv runtimes.
  • [SAFE]: While the skill processes external data (PDB files), it does so within the context of a scientific workflow, and no evidence of command injection or malicious input handling was found.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 11:26 PM
Security Audit — agent-trust-hub — alterlab-rfdiffusion