ddd4j-javalin-data
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external project data including source code, POM files, and database test results. This represents a potential attack surface if the analyzed code contains malicious instructions.
- Ingestion points: External source code files, Maven build configurations (POM), and database test evidence.
- Boundary markers: The skill includes instructions to confirm maintenance lines and specific module locations, but lacks explicit formatting instructions for delimiting untrusted code during analysis.
- Capability inventory: Read-only source code analysis and execution of database round-trip tests via the agent's tool environment.
- Sanitization: No explicit sanitization or escaping of external content before processing is mentioned.
- [METADATA_POISONING]: The file
references/source-evidence.mdcontains a hardcoded local absolute file path (/Users/wandl/workspaces/workspace-ddd4j/workspace-ddd4j-boot/ddd4j-javalin). While not malicious, this is a metadata leak that reveals the author's local directory structure.
Audit Metadata