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
Risk Level
SAFE
Analyzed
Feb 26, 2026, 10:14 AM