diffdock

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The scripts/setup_check.py script uses the __import__ function to verify the presence of required scientific libraries. This is a standard environment check using a hardcoded list of packages (e.g., PyTorch, NumPy, RDKit) and does not represent a risk of arbitrary code execution.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download the official DiffDock repository from GitHub and its corresponding Docker image. These resources are essential for the skill's core functionality and originate from the recognized developers of the tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external research data from CSV and SDF files.
  • Ingestion points: Files are read by prepare_batch_csv.py and analyze_results.py to manage docking tasks and evaluate results.
  • Boundary markers: Boundary markers are not used to separate data from instructions within the files.
  • Capability inventory: The skill has access to file operations and shell execution (Bash) through standard agent tools.
  • Sanitization: Input validation is implemented in prepare_batch_csv.py, which uses regular expressions to ensure that complex names used for file system operations are safe and well-formatted.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 09:46 PM