132-java-testing-integration-testing

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill mandates the execution of shell commands such as ./mvnw compile and mvn clean verify on the local system to validate the project's state. Evidence is found in the prerequisites and safeguards sections of SKILL.md and the reference documentation.\n- [PROMPT_INJECTION]: The skill performs automated analysis of local Java source files to detect infrastructure requirements from import statements, representing an indirect prompt injection surface.\n
  • Ingestion points: Local Java source code and import statements analyzed in Step 1 of the reference file.\n
  • Boundary markers: None specified for the file scanning process.\n
  • Capability inventory: Local shell command execution via Maven and file system write operations.\n
  • Sanitization: No sanitization or validation of the scanned content is described before its use in the agent's configuration logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 05:52 AM