apple-reminders

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the remindctl utility via a Homebrew tap (steipete/tap/remindctl) from a public GitHub repository. This is a standard installation procedure for macOS command-line tools.
  • [COMMAND_EXECUTION]: The skill utilizes the remindctl binary to perform operations such as listing, adding, editing, and deleting reminders. These operations are restricted to the local system's Reminders database.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: While the skill accesses the user's local Apple Reminders data, there are no patterns suggesting unauthorized network transmission or exfiltration of this information. Access to the Reminders database is subject to standard macOS TCC (Transparency, Consent, and Control) permissions.
  • [INDIRECT_PROMPT_INJECTION]: The agent processes output from the remindctl tool (such as reminder titles and list names).
  • Ingestion points: Data enters the agent context when the output of commands like remindctl list or remindctl today is read from the terminal (SKILL.md).
  • Boundary markers: None identified in the provided instructions to delimit tool output from agent instructions.
  • Capability inventory: The skill has the capability to execute shell commands via remindctl (SKILL.md).
  • Sanitization: No explicit sanitization or filtering of the tool's output is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 11:29 PM
Security Audit — agent-trust-hub — apple-reminders