The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill's primary purpose is data augmentation, which involves ingesting and processing external datasets. This creates a surface for indirect prompt injection where malicious instructions could be embedded in training data.
Ingestion points: External datasets processed via the augmentation pipeline.
Boundary markers: Absent; there are no instructions to ignore embedded commands in data.
Capability inventory: Bash(python:*), Bash(pip:*), Write, Edit.
Sanitization: Absent; the skill does not specify validation or sanitization of input data.
[Command Execution] (SAFE): The skill requests Bash(python:*) and Bash(pip:*) tools. While these allow for arbitrary code execution and package installation, no malicious code or suspicious execution patterns were found in the skill's own file.

data-augmentation-pipeline