> For the complete documentation index, see [llms.txt](https://robodex.gitbook.io/robodex-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://robodex.gitbook.io/robodex-docs/frontend/workspace-ui.md).

# Workspace UI

The main UI lives in `apps/web/app/page.tsx`.

It presents:

* project and thread navigation,
* a chat-style prompt input,
* generated robot panels,
* selected robot details,
* simulation checks,
* policy spec creation,
* context summaries.

## Default project behavior

The page creates or loads projects and threads through `workspaceApi`.

When creating projects, the UI can pass a selected agent loop, commonly:

```
creative_qd_v2
```

The backend still validates availability and chooses a fallback when needed.

## Sending a prompt

The prompt flow is:

1. user enters a robot task,
2. `sendPrompt()` calls `workspaceApi.generate(threadId, { prompt, population: 4 })`,
3. backend appends user and assistant messages,
4. backend attaches a `robot_design` thread artifact,
5. frontend reads and displays the generated artifact.

## Robot panel

The robot panel displays the selected generated design using:

* candidate fields,
* render payloads,
* grammar HITL summaries,
* generated artifacts.

It should avoid reconstructing research-loop internals. The backend returns the candidate and payload fields the UI needs.

## Simulation panel

The UI can create simulation specs and run lightweight checks. These are not full RL training runs. The backend records the result and explicitly marks whether full training started.

## Policy panel

The UI can create policy specs describing:

* observations,
* actions,
* reward terms,
* safety constraints.

This is planning/scaffolding for downstream training, not a hidden PPO launcher.

## Context summaries

Context summaries are saved on the thread to make a workspace resumable. They summarize recent prompt, generation, and selected-design state.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://robodex.gitbook.io/robodex-docs/frontend/workspace-ui.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.