discernment-nudge
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFE
Full Analysis
- Operational Logic: The skill functions as a set of supplemental instructions for the AI, guiding it to append follow-up questions to substantive answers. These questions are intended to help users verify facts, probe reasoning, and identify missing context.
- Constraint Management: The skill includes robust criteria for when to remain silent, such as during creative writing, casual conversation, or when the user has already asked for verification. These boundaries prevent the skill from becoming repetitive or interfering with the user's intent.
- Data Handling: No network requests, sensitive file access, or credential management were identified. The skill operates entirely within the text-based interaction context between the AI and the user.
- Output Format: The skill specifies a plain-text output format for its nudges, avoiding the use of complex formatting, HTML, or scripting that could be leveraged for malicious purposes.
Audit Metadata