boxlang-runtime-jsr-223
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill documents the use of
javax.script.ScriptEngine.eval()to execute BoxLang code within a Java Virtual Machine (JVM). This is the primary and intended purpose of the skill, providing a dynamic scripting environment. - Evidence: Code examples utilize
engine.eval()for business logic, rules engines, and template processing. - Mitigation: The documentation explicitly includes a 'Production Checklist' advising users to load scripts only from trusted sources and to use isolated environments for the BoxLang runtime.
- [INDIRECT_PROMPT_INJECTION]: The skill illustrates patterns where scripts are loaded from external files (e.g.,
rules/discount.bxsandtemplates/email.bxs) and executed. This creates a surface where malicious instructions could be introduced if the file source is compromised. - Ingestion points: File system reads using
Files.readString( Path.of( ... ) )inSKILL.md. - Boundary markers: None present in the code examples, but the documentation warns against using untrusted sources.
- Capability inventory: File system access and dynamic script evaluation.
- Sanitization: The skill provides a checklist item specifically for script source validation.
Audit Metadata