testing-essentials

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill uses coercive language to constrain agent behavior.
  • Evidence: The description states the skill is "MANDATORY for ALL test files" and the body instructs the agent to "Follow these with no exceptions."
  • Context: These phrases are typical of prompt injection attacks intended to override system prompts or safety guardrails by asserting absolute priority. In this instance, they are used to enforce a specific Test-Driven Development (TDD) workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 05:51 AM