mcp
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: No malicious behavior, obfuscation, or safety bypass attempts were detected. The skill is a technical implementation guide for developers.
- [COMMAND_EXECUTION]: The skill includes standard shell commands for building and running a Java project using Maven (
mvn spring-boot:run) and verifying API responses viacurl. All network requests are directed tolocalhost, representing standard local development testing. - [DATA_EXFILTRATION]: The documentation explicitly flags that the example MCP endpoints are unauthenticated. This is documented as a 'Security bypass awareness' item in the workflow guide, which provides instructions on implementing proper authentication at the transport level (e.g., using a HandshakeInterceptor) to mitigate unauthorized access risks.
- [INDIRECT_PROMPT_INJECTION]: The skill guides the implementation of tools that process external input through Model Context Protocol arguments. While this establishes an ingestion point for untrusted data, the code templates use structured data handling and type coercion rather than unsafe prompt interpolation, minimizing the risk of injection during data processing.
Audit Metadata