apple-reminders

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the remindctl utility from a public GitHub repository and Homebrew tap as a primary dependency.
  • [COMMAND_EXECUTION]: Executes the remindctl CLI tool to interact with the macOS Reminders database for listing, creating, and modifying entries.
  • [PROMPT_INJECTION]: The skill processes untrusted content from the Reminders database, which creates a surface for indirect prompt injection.
  • Ingestion points: Data returned by remindctl (reminder titles, notes, and list names).
  • Boundary markers: No specific delimiters or instructions are provided to the agent to distinguish data from instructions.
  • Capability inventory: Access to the remindctl CLI for full CRUD operations on Reminders.
  • Sanitization: No input validation or sanitization is performed on the retrieved reminder content.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 12:14 AM
Security Audit — agent-trust-hub — apple-reminders