OUTCOMES
What this Prompt helps you deliver
- The Electron development window runs reliably
- You understand the key directories and restart boundaries
BEFORE YOU START
Before you start
- Root dependencies are installed
- The backend API is reachable
What AI should inspect first
apps/tabtin-electronAGENTS.mddocs/agent/commands.mdREADY TO RUN
Complete Prompt
Copy the complete Prompt with its execution and acceptance boundaries.
Start the TabTin Electron client and help me understand it so I can build my own desktop product. Complete the development startup and distinguish changes that hot reload from changes that require an Electron restart.
Read AGENTS.md, docs/agent/commands.md, and the package configuration in apps/tabtin-electron. Inspect the worktree and development ports such as 5175. Treat the root environment file as the single source; do not create another .env inside the app or stop unrelated Electron processes.
Explain the responsibilities of main, preload, renderer, IPC, and native capabilities and where reusable business logic belongs. Verify API and collaboration health, then launch the client in a persistent terminal. Resolve dependency, native-module, port, or environment errors from real build output. Confirm sign-in, Organization switching, and the main Workspace render against real services rather than fake data. Document the correct reload or restart behavior for renderer, main/preload, IPC, and root environment changes.
Verify the development server, persistent Electron process, and absence of blocking console errors. Run the existing typecheck or lightweight tests and prove one renderer hot update. If main, preload, or IPC changed, perform a real restart and recheck the window.
Report the launch command, ports, process state, directory responsibilities, fixes, backend target, safe shutdown, and recommended next work. Do not commit or push unless I ask.