Documentation

Install and verify the CLI, then prepare capture dependencies in your application.

Choose the creator or reviewer path, then learn how Mockspace organizes a product review.

Install Mockspace, give your coding agent a ready-to-use prompt, and open your first local canvas.

Set up a local canvas, publish a shared review, connect and recover agents, then review and approve work as a team.

Read the packaged skills, preserve existing configuration, and prepare one truthful local flow.

Understand the workflow, setup effort, local and hosted data boundaries, and what to verify in a first team review.

Model one journey as an ordered row that a reviewer can understand end to end.

Describe each step once, then give every meaningful condition a reachable frame.

Use labelled edges for meaningful transitions outside a flow's canonical order.

Make representative application states reachable with fixtures the application owns.

Start the application with its declared environment, validate the spec, and open the live canvas.

Read a shared flow, compare meaningful states, and send a link that preserves review context.

Connect Codex or Claude Code to a deck, track progress, and follow the area being updated.

Supported Mockspace commands for local review, capture, publishing, sign-in, MCP, and workbench automation.

The application-owned config, screen, state, and flow declarations Mockspace loads.

Connect an agent to a local canvas or published review and understand the available context boundary.

Register a project with your authenticated agent or automation, then publish its first review.

How browser sessions, CLI logins, OAuth MCP tokens, machine tokens, and private records differ.

The service route inventory for health, public content, reviews, workbench records, comments, authentication, and MCP.

Diagnose workspace validation, stale app processes, auth walls, capture failures, duplicate states, and access errors.

Turn on the report button in Settings, tell the Mockspace team about a problem with Mockspace itself, and follow it on GitHub.

Use the application's own fixture paths to make important conditions repeatable, coherent, and honest about missing data.

Turn one real journey into a local canvas by describing its screens, making its states reachable, and checking the rendered result.

Put the meaningful conditions of each screen beside the main journey so feedback can address product behavior, not an isolated ideal.

Product improvements and release notes, newest first.