feature-engineering-helper

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION] (MEDIUM): The skill requests Bash(python:*) and Bash(pip:*) permissions. This allows the execution of arbitrary Python scripts and shell-based package management, providing a significant attack surface for local system manipulation.
  • [EXTERNAL_DOWNLOADS] (MEDIUM): Use of Bash(pip:*) allows the agent to download and install any package from the Python Package Index (PyPI). This introduces a risk of installing malicious or compromised third-party libraries (typosquatting/dependency confusion) during the feature engineering process.
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface detected. The skill is designed to ingest and process user-provided data and patterns for feature engineering.
  • Ingestion points: User-provided feature engineering requests and data patterns described in the 'When to Use' section.
  • Boundary markers: None. There are no instructions provided to the agent to distinguish between user data and instructions.
  • Capability inventory: Bash(python:*), Bash(pip:*), Write, and Edit tools allow for code execution, file modification, and environment changes.
  • Sanitization: None detected. The skill does not specify any validation or filtering of the input data before processing it through the allowed tools.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:19 PM