Skills
skills/k-dense-ai/scientific-agent-skills/rdkit/Gen Agent Trust Hub

rdkit

Warn

Audited by Gen Agent Trust Hub on Apr 9, 2026

Risk Level: MEDIUMREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The documentation in 'SKILL.md' recommends using 'pickle.load()' for performance, which is an unsafe deserialization pattern that could allow arbitrary code execution from a malicious file.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via its molecular data processing features. 1. Ingestion points: 'Chem.SDMolSupplier' and 'Chem.SmilesMolSupplier' used in utility scripts. 2. Boundary markers: Absent. 3. Capability inventory: Ability to read and write files locally. 4. Sanitization: The skill performs chemical validation but lacks security-focused input sanitization.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 9, 2026, 10:13 PM