The Agent Skills Directory

Indirect Prompt Injection (SAFE): The skill processes data from an external hardware source (serial port) which constitutes a potential injection surface.
Ingestion points: The monitor_serial.py script reads data directly from a serial connection using serial_conn.readline().
Boundary markers: Output is printed directly to stdout without delimiters or 'ignore instructions' warnings.
Capability inventory: The script supports writing log data to the file system via the --output argument.
Sanitization: Data is decoded as UTF-8 with errors ignored; no specific sanitization for prompt instructions is applied. This risk is inherent to the tool's primary purpose of serial monitoring and is considered a low-risk surface in this context.
External Downloads (SAFE): The skill depends on pyserial and colorama, which are standard, trusted libraries for serial communication and terminal formatting.
Command Execution (SAFE): The script uses argparse for configuration but does not execute external system commands or sub-processes based on user input.

arduino-serial-monitor