alterlab-alphafold-db

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses subprocess.run to interact with the gsutil tool for bulk data downloads. It demonstrates secure coding practices by avoiding shell=True and implementing integer validation for taxonomy IDs to prevent command injection.
  • [EXTERNAL_DOWNLOADS]: Fetches data from the European Bioinformatics Institute (EMBL-EBI) and DeepMind's AlphaFold database. These are well-known scientific sources. It also supports downloading datasets from Google Cloud Storage.
  • [DATA_EXFILTRATION]: No unauthorized data transmission detected. The skill writes scientific data (PDB/mmCIF files) to local directories as part of its primary function.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: Ingests structured JSON and mmCIF metadata from external scientific APIs (EBI).
  • Boundary markers: Data is treated as structured bioinformatics content rather than natural language instructions.
  • Capability inventory: Capabilities are limited to file writing and specific subprocess calls for data retrieval.
  • Sanitization: Includes explicit validation of taxonomy IDs before including them in system commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 12:46 AM
Security Audit — agent-trust-hub — alterlab-alphafold-db