notify-me
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill template uses a bash script to set a variable and run osascript. This approach is susceptible to command injection if the agent interpolates user input containing shell metacharacters (e.g., backticks or subshells) directly into the message variable definition.\n- [DYNAMIC_EXECUTION]: The skill dynamically generates and executes AppleScript code through the osascript utility to interact with the macOS notification system.\n- [INDIRECT_PROMPT_INJECTION]:\n
- Ingestion points: The notification message is derived from user requests or external data ingested by the agent (SKILL.md).\n
- Boundary markers: There are no delimiters or explicit instructions to treat the message content as non-executable data, allowing potential instructions inside the message to be processed by the agent.\n
- Capability inventory: The skill enables access to shell execution and AppleScript system APIs.\n
- Sanitization: The skill documentation does not provide guidelines or mechanisms for sanitizing or escaping the message content before it is processed by the shell environment.
Audit Metadata