paniolo-vitest-test-mock-best-practices

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as a documentation and guidance resource for software testing. All instructions align with legitimate development workflows.
  • [COMMAND_EXECUTION]: The skill defines validation steps using standard project-local commands (npm run test:unit, npm run lint). These commands are scoped to the project environment and do not involve unauthorized execution or privilege escalation.
  • [DATA_EXFILTRATION]: No sensitive data access or exfiltration patterns were identified. The skill correctly identifies that no network access is required for its primary tasks.
  • [PROMPT_INJECTION]: The skill uses instructional language to guide the AI agent's behavior during test generation without attempting to bypass safety filters or override system constraints.
  • [INDIRECT_PROMPT_INJECTION]: While the skill involves processing and testing code that may be provided by a user (Category 8 surface), it mitigates risk by instructing the agent to use established project scripts and existing test utility patterns, which is standard for developer-facing testing tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 03:09 AM
Security Audit — agent-trust-hub — paniolo-vitest-test-mock-best-practices