NYC

docs-cleaner

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE] (SAFE): The skill consists entirely of Markdown content across two files (SKILL.md and a template). No Python, JavaScript, or shell scripts are included.
  • [DATA_EXFILTRATION] (SAFE): No network-related commands, external URLs, or hardcoded secrets were found in the analysis.
  • [PROMPT_INJECTION] (SAFE): No instructions designed to override AI constraints, extract system prompts, or bypass safety filters were detected.
  • [COMMAND_EXECUTION] (SAFE): The skill directs the AI agent to manage files (creation and deletion) as part of a documentation cleanup workflow. This is a high-level instruction for the agent's internal tools rather than a script-based command execution risk. While the skill processes untrusted documentation data (Category 8 surface), the risk is mitigated by the 'Core Principle' of critical evaluation and the instruction to preserve value.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:06 PM