skills/ortus-boxlang/skills/bx-jython/Gen Agent Trust Hub

bx-jython

Warn

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill provides the jythonEval() and jythonEvalFile() functions, which allow for the execution of arbitrary Python code strings and external scripts at runtime.
  • [INDIRECT_PROMPT_INJECTION]: The skill documentation describes a feature where all values in the BoxLang variables scope are automatically bound to the Python engine. This presents an injection surface if the variables are populated with unsanitized data from external sources.
  • Ingestion points: BoxLang variables scope values are automatically mapped to the Python context (referenced in SKILL.md).
  • Boundary markers: None identified in the provided documentation.
  • Capability inventory: Arbitrary Python code execution through jythonEval() and file system access through jythonEvalFile().
  • Sanitization: No sanitization or validation mechanisms are described for the variables being passed into the Python engine.
  • [COMMAND_EXECUTION]: The skill contains installation instructions that involve executing shell commands such as install-bx-module bx-jython and box install bx-jython.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 15, 2026, 09:01 AM
Security Audit — agent-trust-hub — bx-jython