Tickwell is a full-stack app built with Claude Code on Elements.
Keyed issues on a drag-and-drop board and a filtered list, with comment threads, a full change history, assignment emails and CSV import, all live.
Run this app on your machine. Use this demo as inspiration, or as the starting point for your own app. Your agent creates a new app with this one as the scaffold, and you prompt it to add whatever features you want.
Tickwell needed a board and a list that every teammate sees change at once, a history of every edit, email when work is assigned or discussed, team invites and CSV import. Each of those is a part of Elements, so the agent spent its 20 minutes on the tracker itself.
-
A live board and list. Issues are a LiveTable opened per project. Dragging a card is an update through the view, so every open board and list moves with it. Comments and the issue history are LiveTables too.
-
History in one place. The board, the form and the importer all create issues through one function, and every update compares the row before and after inside a transaction, recording each changed field. A database trigger hands out keys such as WEB-14.
-
Email from background jobs. Assignments, comments and invites each schedule a job that sends its email, inside the same transaction as the change.
-
Invites and roles. Admins invite teammates by email through an
@rpc, and every rpc and LiveTable handler checks the signed-in user and their role. -
CSV import as a function call. The import page sends the file to an rpc for a preview, then inserts each row through the project's live view, so open boards fill in as the issues land.
-
Data from SQL files. Two migrations define the tracker and seed three teammates, two projects with 19 issues, their history and comments, and a pending invite.
Elements is an integrated full-stack app framework and a project server. The project server checks the agent's work as it builds: it type-checks the server and browser code, analyzes the program, applies migrations and runs the tests, then reports the results to the agent. A security mistake, such as calling a server-only function from the browser, is a compile error. The agent reads the report and fixes its own mistakes before it tells you a feature is done. The same checks run on every feature you ask for next.