TASTESTUDIO is made up of a Windows desktop app and the Rust Agent. You work on your own PC with the project manager, explorer, Monaco editor, terminal, runs and Git, and the tasks and tests written in the .tastedev definitions of the project folder (TASTEDEV Protocol) are sent through Core to an Agent (Windows · Linux) that meets their requirements. The Agent streams output in real time and leaves results and evidence (logs, browser test records). When something fails, AI only reads: it points to the cause and proposes a fix, and only the changes you approve are applied and tested again. Results can be filed as GitHub issues, and scheduled tests run at the times you set.
From first launch to troubleshooting, written in the order you actually work.
We don't provide ad hoc download links until verified release files and URLs are registered.
Get the Windows installer with a verified code signature and SHA-256 from the Downloads page.
Launch the app, open a project folder and work with the editor, terminal and Git.
Place the Rust Agent on a test machine and connect it to Core, then queue and run the tasks and tests in your .tastedev definitions.
Register folders as projects and pick up where you left off when you reopen them. Folders that have gone missing are found and reconnected.
Open and save multiple tabs with the explorer and the Monaco editor. Files changed outside the app are flagged when you save.
Open a terminal in the project folder, and start and stop programs with run configurations. Standard output and errors are shown separately.
Status, diffs, staging, commits and history inside the app, using the Git installed on your system.
Describe environments, tasks and tests in the project's .tastedev folder. Saved definitions are strictly checked before they appear in the run list.
A standalone runner for Windows and Linux machines. It registers with Core, sends heartbeats, runs the work it receives and streams output in real time.
Work goes into a queue and is assigned to an Agent that has the required capabilities (operating system, tools). Cancellations, timeouts and results at a glance.
Source preparation, build, service start-up, health checks, tests and clean-up run as one pipeline, with the result of every stage recorded.
Run Playwright tests on an Agent and, when they fail, see evidence such as screenshots and traces right in Studio.
AI analyses failed runs read-only and shows proposed fixes as diffs. Only approved proposals are applied, then tested again or reverted.
Draft an issue from a failed run, see similar issues, and on approval file it or link it to an existing issue.
While Core is running, run tests at set times with cron, interval or one-off schedules. Time zones are explicit, and missed times are not run in a burst afterwards.
No public release files have been registered yet. Only ready files appear on the downloads page.
The docs only list environments verified for each product. “Being verified” means support isn't confirmed yet.
The key features in these docs and the product page are written based on the current project source.