backend-unit-test
Pass
Audited by Gen Agent Trust Hub on Mar 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to construct and execute shell commands (
mvn compiler:testCompile,mvn surefire:test) using parameters such as<module>and测试类名(test class name) derived from user input. Direct interpolation of untrusted strings into shell commands without escaping or validation presents a risk of command injection. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted user input to determine command-line arguments.
- Ingestion points: User requests specifying modules, test classes, or functional paths (e.g., in
SKILL.mdexamples). - Boundary markers: Absent. The skill does not use delimiters or instructions to treat user input as data only.
- Capability inventory: The skill has the capability to execute shell commands (
mvn), find files (find), and create new Java/YAML files on the file system. - Sanitization: Absent. There are no instructions for the agent to validate, escape, or sanitize the user-provided strings before using them in shell commands or file paths.
Audit Metadata