reminders
Warn
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download and install a third-party CLI tool (
rem) directly from a personal GitHub repository usinggo install github.com/BRO3886/rem/cmd/rem@latest. - [REMOTE_CODE_EXECUTION]: Executing a remote installation command (
go install) from an unverified source at runtime allows for the execution of arbitrary code if the repository or build process is compromised. - [COMMAND_EXECUTION]: The skill performs various shell operations to interact with the macOS environment, including executing the
remtool for reminder management and thedatecommand for temporal calculations. - [PROMPT_INJECTION]: There is a surface for indirect prompt injection because the agent processes data (titles, notes, and list names) retrieved from the user's Reminders app.
- Ingestion points: Data retrieved from the system via
rem search,rem list, andrem showcommands. - Boundary markers: Absent. The instructions do not specify any delimiters or warnings to ignore commands embedded within the reminder text.
- Capability inventory: The skill can execute shell commands, manage files (reminders), and potentially perform network operations if instructed by malicious data.
- Sanitization: Absent. The skill reads and acts upon the retrieved reminder content without validation or filtering.
Audit Metadata