The Agent Skills Directory

Indirect Prompt Injection (HIGH): The skill is designed to read data from external Excel files and provide the raw output to the agent, creating a significant attack surface for indirect prompt injection.
Ingestion points: scripts/read_xlsx.py and scripts/xlsx_to_json.py ingest content from .xlsx files using the openpyxl library.
Boundary markers: Absent. The scripts output content as raw Markdown or JSON text without using delimiters or system-level instructions to ignore embedded commands.
Capability inventory: The agent is explicitly instructed to execute shell commands (wsl python3 ...) and is encouraged to use 'Write' tools to save the resulting data to the filesystem.
Sanitization: Absent. The Python scripts only perform basic string conversion and whitespace stripping (str(value).strip()), failing to escape Markdown syntax or filter for instructional language.
Command Execution (MEDIUM): The skill's primary function relies on the agent executing arbitrary shell commands via WSL. If the agent fails to strictly validate the user-provided file path, this could be abused to target sensitive files on the host system or execute unintended scripts.
External Downloads (LOW): The skill documentation instructs users to install the openpyxl package from PyPI (pip install openpyxl). While this is a trusted and standard library, the installation of third-party dependencies is an external requirement for the skill's operation.

xlsx-reader