steering-user-elicitation

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The skill consists entirely of Markdown instructions defining how an AI agent should interact with users. It does not contain any scripts, binaries, or automated configuration steps.
  • [INDIRECT_PROMPT_INJECTION]: While the skill defines how to process untrusted user input (data analysis requests), its primary function is to implement a defensive communication strategy (elicitation). It instructs the agent to seek clarification, verify metrics against a data dictionary (../reading-data-dict/), and avoid making assumptions, which serves to harden the agent against ambiguous or potentially malicious data queries.
  • [SAFE]: No sensitive file access, network operations, obfuscation, or persistence mechanisms were detected. The skill uses standard instructional language to guide AI behavior towards better requirement elicitation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 09:38 AM
Security Audit — agent-trust-hub — steering-user-elicitation