Your workspace and setup together
The homepage now brings project work and setup into one layout. Returning readers can choose a project, continue a draft, and open the canvas. New readers can explore the Fieldnotes example before preparing their own app. The example is synthetic, with three screens and seven states you can inspect without an account.
On smaller screens, the homepage preview scrolls horizontally so the product stays readable. Open the interactive example to move between screens, select empty and error states, and try their recovery actions.
A clearer first local review
Quickstart follows three steps: install Mockspace, paste the setup prompt into your coding agent, and open the local viewer URL it returns. Copy buttons are available for commands and agent prompts. If clipboard access is unavailable, the text is selected for manual copying.
The New project page helps customize that setup prompt for an existing checkout or an explicitly chosen new clone. It does not create a hosted project just by generating a prompt. A project appears on the shared desk when a review is published. Start with the Quickstart, or read the agent guide for the detailed procedure.
Public guides and a consistent reading layout
The documentation covers flows, screens, states, local development, sharing, and the supported interfaces. Search finds published guides and commands. Article widths stay consistent as you move between pages, while long command blocks scroll within their own area.
The shared header connects the workspace, docs, and blog. Appearance stays in a compact control with System, Light, and Dark choices. The primary setup action remains available on phones.
Evaluate with a real journey
The team evaluation guide explains what the workflow needs, what remains local, and what changes when you publish a review. It includes a small evaluation sequence your developers, designers, and product reviewers can run together.
The best first test is one representative journey with visibly different states. Ask your agent to report unreachable states and missing fixtures, then judge whether the canvas makes the review easier to complete. A rendered example is useful evidence; it should also make clear which work still needs to happen in your own app.