clarify-first
Pass
Audited by Gen Agent Trust Hub on Feb 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a structured alignment protocol that requires AI agents to pause and clarify ambiguous instructions, reducing the risk of unintended behavior.
- [SAFE]: It introduces a weighting system for technical assumptions, where environment or dependency-related assumptions trigger an immediate request for clarification.
- [SAFE]: The skill mandates a two-phase execution process (Plan then Code) for medium to high-risk tasks, incorporating human-in-the-loop safety checks and mandatory rollback preparation for high-impact actions.
- [SAFE]: No evidence of data exfiltration, credential exposure, or unauthorized network operations was found. The skill is composed entirely of instructional markdown and localized templates.
Audit Metadata