ddd4j-cloud-context
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze external data, including Project Object Model (POM) files, source code, and test evidence. This creates an attack surface for indirect prompt injection if the analyzed files contain hidden instructions. The skill mitigates this by including explicit safety rules against outputting sensitive data or real tenant information, and using fictional data for examples.
- [COMMAND_EXECUTION]: The skill documentation suggests the use of
npx skills addcommands to install additional modules (e.g.,ddd4j-cloud-tenant,ddd4j-cloud-feign) from thefull-stack-skillsvendor. These commands are part of the intended workflow for managing related skills within the vendor's ecosystem. - [DATA_EXPOSURE]: The file
references/source-evidence.mdcontains a hardcoded absolute local filesystem path (/Users/wandl/workspaces/workspace-ddd4j/...). This reveals the author's local directory structure and username, which is a minor metadata exposure, but it is used legitimately to define the workspace root for the agent's analysis tasks.
Audit Metadata