NYC

networkx

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
  • [Dynamic Execution] (MEDIUM): Documentation for pickle.load() and nx.read_gpickle() in references/io.md represents a potential security risk.
  • Evidence: The file references/io.md explicitly demonstrates how to use pickle.load(f) and nx.read_gpickle('graph.gpickle').
  • Risk: If an agent or user follows these examples to load a graph file from an untrusted external source, it could lead to arbitrary code execution (ACE) because the Python pickle module can execute arbitrary code during deserialization.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:23 PM