The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes dynamically generated AppleScript code using the osascript utility via a Bash tool. This capability is used to both query existing events and create new ones within the macOS Calendar application.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) because it processes untrusted data from user text and images without sufficient sanitization. ● Ingestion points: User-provided text, chat screenshots, posters, or email screenshots are used to extract event metadata in Step 2. ● Boundary markers: The skill does not define delimiters or specific 'ignore' instructions for the extracted content before inserting it into the AppleScript command string. ● Capability inventory: The skill can perform arbitrary AppleScript operations through the osascript interface, which has access to system data and applications. ● Sanitization: There are no instructions to escape or validate user-provided strings (like 'summary' or 'description') before they are interpolated into the make new event AppleScript template, potentially allowing an attacker to 'break out' of the string and execute unintended AppleScript commands.

ljg-calendar