The Agent Skills Directory

[SAFE]: The skill implements a robust security model for handling user data backups, focusing on integrity and preventing data corruption.
Employs SHA-256 checksums (implemented in pure TypeScript for portability and transparency) to verify backup files before restoration.
Uses strict schema validation via the Valibot library to ensure that restored data conforms to expected formats.
[SAFE]: Includes defensive mechanisms against CSV injection attacks by neutralizing common formula leaders in exported files.
Text cells starting with =, +, -, @, \t, or \r are prepended with an apostrophe to prevent execution in spreadsheet software.
[SAFE]: Protects application-level security and revenue models by excluding sensitive system tables.
Explicitly filters out entitlements (purchases) and error_log (diagnostics) from backup payloads, ensuring these cannot be tampered with by users via file editing.
[SAFE]: The provided utility scripts (check-backup.mjs and inspect-backup.mjs) are static analysis tools designed to run in a developer's local environment. They do not perform network requests or execute arbitrary external code.

data-export-and-restore