bx-ai-pipelines

Pass

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and serves as a technical reference for the BoxLang AI pipeline system. All provided code examples demonstrate legitimate usage of the framework's Built-in Functions (BIFs) and vendor-specific modules.
  • [PROMPT_INJECTION]: The skill documents the use of ${variable} and ${_input} placeholders for prompt interpolation in SKILL.md. This represents a surface for indirect prompt injection where untrusted data (e.g., user-supplied text or prior AI outputs) enters the model context. While standard for templating frameworks, developers are responsible for implementing boundary markers or sanitization for these inputs.
  • [REMOTE_CODE_EXECUTION]: The documentation demonstrates the use of aiTransform() with closures (e.g., r -> r.content) to process AI responses. In the context of the BoxLang ecosystem, these closures represent local functional transformations for data mapping and deserialization rather than untrusted remote code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 15, 2026, 02:44 PM