start
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill initializes a local productivity system by checking for and creating necessary markdown files (TASKS.md, CLAUDE.md) and a directory (memory/). All operations are restricted to the local working directory.
- [SAFE]: The skill requests access to sensitive user data (tasks, emails, messages) specifically for the purpose of building a local 'memory' of projects and people. There are no network-enabled tools or commands that would allow for data exfiltration.
- [SAFE]: The instructions explicitly forbid the use of potentially dangerous shell commands like
openorxdg-open, instead directing the user to open files manually in their browser. - [SAFE]: Tool access is restricted to benign file-searching and reading utilities (Read, Glob, Grep).
Audit Metadata