python-sdk

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The documentation includes code examples using the eval() function to implement a calculator tool. While intended for demonstration, using eval() on unsanitized input is a significant security risk (command injection) if developers copy this pattern into production environments.
  • [COMMAND_EXECUTION]: The SDK provides a code_execution capability for agents, allowing them to dynamically generate and run code. This is a core platform feature that requires strict permission management and oversight of agent autonomy.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to install the inferencesh package from PyPI, the standard package registry for Python, which is the expected delivery method for the vendor's official SDK.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 09:11 AM