The Playground provides a unified, interactive environment where users can work with prompt templates within a single workspace. It exists to allow controlled experimentation, validation, and comparison of prompts without switching screens or losing context. The Playground is primarily used by users who design, test, and analyze prompts, including Admins and Members, while the system maintains execution results and metrics tied to each prompt version.
A product engineer opens the Playground to evaluate a content-generation prompt before deployment. They remain on the same screen while editing the prompt, running it against a selected model, reviewing outputs, and examining performance metrics, allowing them to understand prompt behavior without leaving the screen.
This section establishes which prompt template and version are currently active in the Playground. It exists to ensure that all actions, such as editing, running, analysis, and testing, are scoped to a clearly identified template version. Users rely on this context to maintain traceability, accountability, and consistency across experimentation activities.
A team member selects a marketing copy template labeled as version 1.0 and sees its risk level, description, and author information displayed. All subsequent edits and test runs remain associated with that version, allowing the team to later understand exactly which template produced specific results.
The Editor tab provides a focused space for authoring and modifying prompt content before execution. It exists to allow users to refine prompt wording, structure, and intent while maintaining visibility into prompt length and variables. This tab is primarily used by users responsible for prompt design and iteration.
A content strategist uses the Editor tab to adjust the wording of a blog-generation prompt, reviewing the text directly and confirming that it reflects the intended tone and scope before testing it with real inputs.
The Analytics tab provides visibility into performance metrics for all executions related to the selected prompt version. It exists to help users understand trends in efficiency, costs, and usage over time.
The A/B Testing tab enables comparison of different prompt versions within the Playground. It exists to support evidence-based decision-making by showing which version performs better under similar conditions.
