Can AI replace Timely?

Build a closest honest personal substitute for Timely in an empty repository.
Use Tauri 2, React, TypeScript, SQLite, and optional OS accessibility APIs; do not offer alternative stacks.
The core loop is: record opt-in app names locally, turn them into suggested time entries, review patterns locally, and export a transparent activity record.
Make the first run work locally with one documented command.
Store all user data locally by default and make export straightforward.
Put secrets in .env, ship .env.example, and never commit credentials.
Create one-click start and stop, manual entries, idle detection, tags, projects, and notes.
Keep any application-name tracking opt-in and never record window contents or keystrokes.
Add Pomodoro-style focus sessions, break reminders, and an optional allowlist blocker.
Provide daily, weekly, project, and category reports with editable corrections.
Export CSV and JSON and make the underlying event log inspectable.
Add automatic local backups and a privacy screen explaining every collected field.
Include clear empty, loading, success, and recoverable error states.
Add input validation, safe filenames, and graceful handling of unavailable APIs.
Write focused tests for the core transformation and one end-to-end happy path.
Create a README with setup, architecture, permissions, data location, and backup steps.
Do not add accounts, billing, telemetry, analytics, or a hosted control plane.
Do not claim to reproduce proprietary data, network liquidity, regulated access, or frontier infrastructure.
Deliberately leave out employee surveillance.
Deliberately leave out payroll and billing automation.
Deliberately leave out cross-device blocking and behavioral coaching services.
Finish by running the tests and listing the exact commands used.