ddd4j-boot-extensions
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill supports the integration of the QLExpress extension, which executes user-supplied expressions, creating an inherent attack surface for indirect prompt injection.
- Ingestion points: Untrusted script content and expressions handled by the QLExpress submodule as described in
SKILL.mdandexamples/02-extensions-review.md. - Boundary markers: The skill does not define specific delimiters for separating untrusted script content from agent instructions, though it warns against loosening security during assembly.
- Capability inventory: The extension allows for the wiring and runtime execution of arbitrary logic via the QLExpress engine, which is a high-capability execution tool.
- Sanitization: The skill includes warnings that configuration can leak credentials and that script execution is sensitive, but it lacks specific guidance on sanitizing or validating the ingested scripts.
Audit Metadata