NYC

coding-level

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWNO_CODE
Full Analysis
  • [PROMPT_INJECTION] (SAFE): The skill uses instructional modifiers to define the agent's response persona. While it includes a section labeled 'IMPORTANT' regarding task planning, these instructions are transparent, benign, and intended to improve the quality of the agent's work rather than bypass safety controls.
  • [DATA_EXFILTRATION] (SAFE): No network operations or sensitive file access patterns were detected. The skill only references a local configuration file for state management.
  • [COMMAND_EXECUTION] (SAFE): There are no shell commands, subprocess calls, or scripts included in this skill.
  • [NO_CODE] (INFO): The skill is entirely markdown-based, functioning as a system prompt extension rather than a functional tool or extension with executable logic.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 01:31 PM