ddd4j-cloud-bom

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [DATA_EXFILTRATION]: The file references/source-evidence.md contains a hardcoded local file path (/Users/wandl/workspaces/workspace-ddd4j/workspace-ddd4j-cloud/ddd4j-cloud) which discloses internal directory structures.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data in the form of Maven POM files and CI output. 1. Ingestion points: The skill reads cloud-parent, cloud-dependencies, and cloud-bom POM files as described in SKILL.md and references/workflow.md. 2. Boundary markers: No specific delimiters or instructions to ignore embedded commands within POM files are present. 3. Capability inventory: The skill utilizes file system read access, build execution (mvn), and invokes sub-tools (ddd4j-cloud-version-selection). 4. Sanitization: There is no evidence of sanitization or filtering applied to the content of the POM files before processing.
  • [EXTERNAL_DOWNLOADS]: The documentation mentions installation of additional skills via npx skills add full-stack-skills/ddd4j-skills. These are identified as vendor-owned resources for the 'full-stack-skills' author.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 04:44 AM