datamol-cheminformatics

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the datamol library and several optional dependencies (s3fs, gcsfs, adlfs, Pillow, cairosvg, openpyxl) via uv pip install to enable core functionality and cloud storage support.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it is designed to ingest and process data from external, potentially untrusted sources.
  • Ingestion points: Data enters the agent's context through file reading functions like dm.read_sdf, dm.read_csv, dm.read_excel, and the auto-detecting dm.open_df (documented in SKILL.md).
  • Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore potentially malicious text embedded within the molecular data fields.
  • Capability inventory: The skill performs file system I/O, network operations for remote files (S3, GCS, HTTP), and parallel task execution across all processing scripts.
  • Sanitization: The skill utilizes domain-specific chemical standardization (dm.standardize_mol), but lacks general input sanitization for non-chemical data fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 12:54 AM