notes-export
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local TypeScript file (
scripts/notes.ts) using thebunruntime to automate the export of data from the macOS Notes database via Apple Events. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Apple Notes. Malicious content within these notes could potentially influence the agent's behavior during the export process or affect downstream skills that analyze the resulting markdown files.
- Ingestion points: Note content and metadata from all accounts and folders in Apple Notes, accessed via the
notes.tsscript. - Boundary markers: The skill uses YAML frontmatter to separate metadata in exported files but lacks instructions for the agent to ignore instructions embedded within the note body during processing.
- Capability inventory: The skill is configured with
BashandReadtool access. - Sanitization: Implements filename sanitization (
safeFilename) to prevent path traversal or collision, but does not sanitize the note body content for potential prompt injection triggers. - [DYNAMIC_EXECUTION]: The instructions designate the skill as "Self-Evolving," explicitly directing the agent to modify its own configuration (
SKILL.md) and logic (scripts/notes.ts) to resolve bugs or behavioral drift. This represents a self-modifying code pattern where the agent is authorized to rewrite its own executable instructions.
Audit Metadata