apple-notes

Pass

Audited by Gen Agent Trust Hub on Jul 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill interfaces with the macOS Notes application by using subprocess.run to call osascript. It avoids potential script injection by passing dynamic user data (such as queries and note content) via environment variables, which are then accessed within AppleScript using the system attribute command.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it processes user-provided note content. 1. Ingestion points: Note titles and bodies retrieved from the local Notes.app in scripts/notes.py. 2. Boundary markers: Note content is returned to the agent without explicit delimiters or warnings to ignore embedded instructions. 3. Capability inventory: The skill can modify or delete notes and write to the local file system. 4. Sanitization: Input content is escaped via html.escape and output is converted from HTML to Markdown.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 4, 2026, 03:22 PM
Security Audit — agent-trust-hub — apple-notes