The Agent Skills Directory

PROMPT_INJECTION (LOW): The skill is vulnerable to indirect prompt injection (Category 8) as it processes data from the macOS Reminders database. Ingestion points: Data entering the context via the list-reminders command output. Boundary markers: Absent; there are no instructions to the agent on how to delimit or ignore instructions found within reminder content. Capability inventory: Uses osascript for system interaction, python3 for processing, and writes to logs/reminders.log. Sanitization: Absent; while input parameters are validated, the content of the reminders themselves is not sanitized before being returned to the agent.
COMMAND_EXECUTION (LOW): The skill executes a local shell script (reminders.sh) and uses osascript. These tools possess significant system permissions, but are utilized for the primary task of reminder management. Data is passed via stdin to prevent sensitive information from appearing in process lists.

macos-reminders