automating-notes
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements legitimate macOS automation tasks for Apple Notes using JXA and the PyXA library. The provided scripts for creating, searching, and organizing notes align with the stated purpose and follow established platform conventions.
- [COMMAND_EXECUTION]: The skill includes Python and Shell setup scripts (scripts/set_up_notes_automation.py and scripts/set_up_notes_automation.sh) that execute AppleScript commands via osascript. These commands are static, hardcoded, and designed only to trigger standard macOS automation permissions for the Notes app, posing no security risk from external input.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing user-supplied text and HTML to create notes. It addresses potential risks by providing explicit instructions for the agent to sanitize this input and avoid dangerous tags like or event handlers. 1. Ingestion points: The create_note.py script and JXA examples accept content arguments for note bodies. 2. Boundary markers: The skill relies on natural language instructions for the agent to validate and sanitize input rather than programmatically enforced delimiters. 3. Capability inventory: The skill is scoped to Apple Notes automation and does not possess network access or generic file system write capabilities beyond standard Notes functionality. 4. Sanitization: The documentation recommends stripping tags like and from note bodies before creation.
Audit Metadata