ddd4j-cloud-feign
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from the user's local workspace, including build configuration files (POMs), source code, and test execution results as defined in references/source-evidence.md. This creates an attack surface where malicious code or instructions within the analyzed project could attempt to influence the agent's behavior.
- Ingestion points: Local file system access to /Users/wandl/workspaces/workspace-ddd4j/workspace-ddd4j-cloud/ddd4j-cloud and associated source/test files.
- Boundary markers: The instructions explicitly include safety markers in the 'Privacy and Security' section, directing the agent to never output credentials and to use desensitized fictional values for examples.
- Capability inventory: The skill reviews Feign client configurations, interceptors, and retry policies, providing implementation recommendations based on the analyzed source code.
- Sanitization: Mandatory redaction of Nacos, Redis, broker, and database credentials is required for all outputs.
- [DATA_EXPOSURE]: The skill references a specific local workspace path (/Users/wandl/workspaces/workspace-ddd4j/workspace-ddd4j-cloud/ddd4j-cloud) for locating source evidence. While this path appears to be an environment-specific reference, the skill explicitly prohibits the exfiltration of sensitive data, including real tenant IDs or private repository credentials, mitigating the risk of accidental exposure during the analysis process.
- [EXTERNAL_DOWNLOADS]: The skill mentions the installation of related capability modules via 'npx skills add full-stack-skills/ddd4j-skills'. These references target the official repository of the skill author ('full-stack-skills') and are documented as standard extension mechanisms for the ddd4j development environment.
Audit Metadata