rowan

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the rowan-python library, which is the official client for the Rowan service.
  • [COMMAND_EXECUTION]: The documentation provides standard CLI commands for environment setup, including package installation via uv pip and setting authentication credentials via export.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external molecular data such as SMILES strings and PDB files which are processed by the agent and the Rowan API.
  • Ingestion points: Functions like stjames.Molecule.from_smiles, rowan.upload_protein, and rowan.create_protein_from_pdb_id (SKILL.md, proteins_and_organization.md).
  • Boundary markers: Data is passed through structured API calls and specialized library objects (Molecule, Protein).
  • Capability inventory: The skill performs network requests to Rowan's API endpoints and supports local file writes for simulation results like PDB and SDF files.
  • Sanitization: The skill provides a protein.sanitize() method designed to clean and standardize protein structures, reducing the risk of processing malformed or malicious data artifacts (proteins_and_organization.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 02:49 PM