ask-user-question

Pass

Audited by Gen Agent Trust Hub on Mar 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or security vulnerabilities were detected in the skill's source code or configuration.
  • [DATA_EXFILTRATION]: The skill communicates with a local server (localhost:9227) to facilitate the user interface. This is standard behavior for MCP tools that bridge to a local GUI and does not represent an exfiltration risk.
  • [REMOTE_CODE_EXECUTION]: The skill uses standard, well-known dependencies and does not employ any dangerous execution patterns such as eval, exec, or downloading remote scripts.
  • [PROMPT_INJECTION]: While the skill returns user-provided text to the agent, which is an indirect injection surface, this is the intended and necessary functionality for a user-input tool and occurs within the local trust boundary.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 2, 2026, 12:14 PM