521-frameworks-micronaut-testing-unit-tests
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to execute shell commands such as
./mvnw compileandmvn clean verify. While these are standard Maven commands for Java development, they provide a vector for executing untrusted code if the project build configuration is malicious.\n- [INDIRECT_PROMPT_INJECTION]: The skill involves reading and refactoring local source code and project configurations, which serve as untrusted inputs that could contain malicious instructions for the agent.\n - Ingestion points: The agent reads local Java source files and Maven build configuration files (pom.xml) to assess project context.\n
- Boundary markers: There are no explicit instructions or delimiters defined to help the agent distinguish between its instructions and the untrusted project data being analyzed.\n
- Capability inventory: The skill workflow grants the agent the ability to write to the file system for refactoring and to execute shell commands for build verification.\n
- Sanitization: The skill lacks validation or sanitization protocols for the source code and configurations it ingests before processing them.
Audit Metadata