401-frameworks-quarkus-core
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill explicitly requires running
./mvnw compileand./mvnw clean verifyto validate Quarkus application structure and changes. This involves executing local binaries and project-defined build logic, which is standard behavior for development and build-related tasks. - [DYNAMIC_EXECUTION]: By instructing the agent to run Maven build commands, the skill initiates the compilation of Java source code. This is categorized as a low-risk dynamic execution pattern when used for standard build processes on local source code.
- [INDIRECT_PROMPT_INJECTION]: The skill processes and refactors Java source files and configuration properties from the project, which creates a theoretical ingestion surface for adversarial instructions hidden within the source code.
- Ingestion points: The skill reads project-specific Java files and application.properties.
- Boundary markers: No specific delimiters or boundary instructions are provided to the agent to distinguish between its instructions and the code data it processes.
- Capability inventory: The skill possesses the ability to execute shell commands (via Maven) and write modifications to the file system.
- Sanitization: No explicit sanitization or filtering of the processed code content is mentioned in the guidelines.
Audit Metadata