Playground Evaluation Workbench Announcement Copy

June 9, 2026

We rebuilt the Agenta Playground from scratch. It is now the main workspace for your AI applications.

Two major additions came with it.

You can attach evaluators to the Playground and score outputs as you edit. Change the prompt, run again, and the scores update next to each output.

You can also load test sets into the Playground. This lets you work on your evals and add test cases directly from the Playground.

X / Twitter

We rebuilt the Agenta Playground from scratch. It is now the main workspace for your AI apps.

Attach evaluators to score outputs as you edit, and load test sets to work on your evals and add test cases without leaving the Playground.

1/ We rebuilt the Agenta Playground from scratch. It is now the main workspace for your AI applications.

2/ You can attach evaluators to the Playground and score outputs as you edit. Change the prompt, run again, and the scores update.

3/ You can also load test sets. Work on your evals and add test cases directly from the Playground.

We rebuilt the Playground from scratch. It is now the main workspace for your AI applications.

Two things you can do now:

Attach evaluators to score outputs as you edit.
Load test sets to work on your evals and add test cases directly from the Playground.