* refactor(web): rename Execution mode and tighten settings dialog UI - Rename "Settings → Execution & model" to "Settings → Execution mode" across the web UI, i18n keys, docs, and e2e selectors. - Redesign SettingsDialog: kicker + title row in the modal head, a flatMap-driven agent grid that renders the inline test-result row beside the selected card, compact unavailable cards with right-aligned install/docs links, and an install guide that only shows when the user has no working agent picked. - Trim verbose subtitle / hint copy across chat model, CLI proxy, media providers, custom instructions, and memory sections. - Add an `info` Icon variant for the redesigned settings hints. - Update e2e selectors and docs that referenced the old menu label. Co-authored-by: Cursor <cursoragent@cursor.com> * refactor(web): polish Settings dialog — media providers, skills, MCP Media providers - Hide internal Stub fixture provider (settingsVisible: false) - Split provider list into Available (integrated, editable) and Coming Soon (collapsed <details> drawer with name/hint/Docs link only) - Drop right-side Integrated/Configured badges from every row; all rows in the main list are integrated by definition; inline grey "Saved" chip next to the provider name is the only status indicator now - "Saved" badge moves inline to the right of the provider name and uses a neutral grey treatment (was a standalone green pill below the name) - "Reload from daemon" button shows a 2s green "✓ Reloaded" flash on success instead of leaving a permanent paragraph under the header; errors remain sticky Skills - Replace three pill-row filter banks (Source, Type, Category) with a compact single-row toolbar: search + three inline <select> dropdowns side by side; active filter highlighted with a stronger border MCP server - Shorten section hint to one line - Move WHAT YOUR AGENT CAN DO capabilities above the client dropdown (motivate before asking to act) - Move "Build the daemon first" warning below the code block where it contextually explains why the command might fail, not as a top-level error before the user has done anything - Downgrade "Restart your client" left-border from accent orange to border-strong grey — it is a next step, not a warning External MCP - Shorten section hint to one line Misc CSS - Add .sr-only utility for accessible off-screen live regions - Add button.ghost.is-success-flash for transient success feedback - Add .library-filter-selects / .library-filter-select for dropdown filter rows - Add .media-provider-coming-soon-* for the roadmap drawer Co-authored-by: Cursor <cursoragent@cursor.com> * [codex] Add Cursor Agent auth diagnostics (#1538) * Add Cursor Agent auth diagnostics * Handle Cursor not logged in auth status * Address Cursor auth review feedback * Classify Cursor stdout auth failures * test: expand Memory and Routines coverage (#1521) * test: expand settings and packaged coverage * test: extend memory settings coverage * test: cover routine settings failure states * test: cover routine operation failures * test: fix daemon test typing on CI * test: decouple packaged smoke from orbit bug * test: avoid live memory LLM calls in route tests * test: fix daemon fetch typing in CI * fix: restore preview comment and inspect toggles * test: align manual edit flow with current inspector UX * test: align comment attachment flow with current preview comments UI * fix: probe resolved Codex launch path during detection * fix: remove duplicate board activation helper after rebase * test: update ghost cli detection mock * test: align FileViewer toolbar expectation * ci: move full app tests to extended lane * ci: run app tests by changed scope * ci: cover shared app inputs in test scopes * ci: avoid setup-node cache in windows packaged smoke * test: align extended settings and manual edit flows * refactor(web): rename Execution mode and tighten settings dialog UI - Rename "Settings → Execution & model" to "Settings → Execution mode" across the web UI, i18n keys, docs, and e2e selectors. - Redesign SettingsDialog: kicker + title row in the modal head, a flatMap-driven agent grid that renders the inline test-result row beside the selected card, compact unavailable cards with right-aligned install/docs links, and an install guide that only shows when the user has no working agent picked. - Trim verbose subtitle / hint copy across chat model, CLI proxy, media providers, custom instructions, and memory sections. - Add an `info` Icon variant for the redesigned settings hints. - Update e2e selectors and docs that referenced the old menu label. Co-authored-by: Cursor <cursoragent@cursor.com> * refactor(web): settings dialog UX polish — layout, dedup, and interactions - Remove duplicate section headers from all settings sections (Notifications, Appearance, Privacy, About, Design Systems, Skills, MCP server, Connectors, Media providers, Routines) - Restructure Notifications cards: title + toggle on same row, hint below - Restructure Skills toolbar: search + New skill button in row 1, filter dropdowns in row 2 with left-aligned labels - Restructure Pet section: tabs and Wake button on same row - MCP server: group capabilities and setup into separate cards, remove nested double border on client picker - Connectors: show connect errors as toast instead of inline card text, position toast inside panel, hide single-provider tab - Media providers: move Reload button to left-aligned small ghost button - Memory: info icon shows path on hover, Path copied badge inline; Extraction history and MEMORY.md as standalone collapsible cards; group header hidden when only one type visible - Pet grid cards: Adopt button hidden until hover, icon-only when adopted, description truncated to 2 lines, text fills full width via abs positioning - Agent cards: selected state uses accent border only, no background change - Add sun/moon icons to Appearance theme buttons (Light/Dark) - Shorten several hint strings for clarity Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web): resolve i18n review comments from PR #1568 - Update settings.title and settings.envConfigure to localized "Execution mode" in all 17 non-English locale files - Add settings.memoryFlashPathCopied to all locales and use t() in MemorySection instead of hardcoded English "Path copied" - Add settings.agentModelHead to all locales and use t() in SettingsDialog for "Model for:" agent model row header Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web): update tests to match settings dialog redesign - Add role prop to Toast (alert/status) so error toasts from ConnectorsBrowser are announced immediately by screen readers - Clear connectErrorToast on successful connector retry - Update SettingsDialog.execution tests: - Remove heading assertions for About and MCP server (headers were intentionally removed as duplicate nav labels) - Rewrite CLI env test to use codex-only fields (per-agent filtering means only selected agent's fields are shown) - Update Composio key hint text assertion to match shortened copy - Replace filter button click with select change for Type filter - Replace Configured/Unsupported/Integrated badge checks with updated assertions matching the new media provider UI - Replace disabled BFL row test with coming-soon section check - Update SettingsDialog.media test: remove Fal.ai input assertions (non-integrated providers no longer have editable fields) Co-authored-by: Cursor <cursoragent@cursor.com> * fix(web): unblock CI for #1568 Three small fixes to get Playwright back to green on the settings dialog redesign: 1. `en.ts`: revert `settings.envConfigure` to "Configure execution mode". This PR collapsed both `settings.title` (header gear) and `settings.envConfigure` (entry-side foot pill) to the same string "Execution mode", so `getByRole('button', { name: 'Execution mode' })` resolved to two elements and tripped Playwright strict mode in the three Composio-flow tests (entry-configuration-flows.test.ts:174, 228, 285). Restoring the distinct label also gives screen readers a clearer hint for the pill, which doubles as a status display. Non-English locales still alias the two keys; happy to follow up on those, but they don't gate the (English-only) Playwright suite. 2. entry-configuration-flows.test.ts:167 — `Connectors` heading is now rendered at `<h2>` in the modal-head (SettingsDialog.tsx:1545), with the inner `<h3>` removed by design (see comment around line 1448). Updated the assertion from `level: 3` to `level: 2`. 3. project-management-flows.test.ts:360 — same change for the `Pets` heading. Verified locally with `pnpm --filter @open-design/web typecheck` and `pnpm --filter @open-design/e2e typecheck`. The actual Playwright specs need the dev server up; I didn't rerun them here, but the locator changes are mechanical and match the new DOM. * fix(web): use exact match for Execution mode button locator Playwright's `getByRole({ name })` defaults to substring matching, so `{ name: 'Execution mode' }` still resolved to both the header gear (aria-label "Execution mode") and the entry-side foot pill (aria-label "Configure execution mode" — substring contains "Execution mode"). Strict mode tripped in the three composio-flow tests at lines 202, 257, and 319. Adding `exact: true` makes each call resolve to just the header gear, which opens the same dialog the foot pill does — the test outcomes are unchanged. --------- Co-authored-by: chaoxiaoche <chaoxiaoche@chaoxiaochedeMacBook-Pro.local> Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: Caprika <56862773+alchemistklk@users.noreply.github.com> Co-authored-by: shangxinyu1 <shangxinyu@refly.ai> Co-authored-by: lefarcen <935902669@qq.com>
18 KiB
Quickstart
English · Português (Brasil) · Deutsch · Français · 日本語 · 简体中文 · 繁體中文
Run the full product locally.
Environment requirements
- Node.js:
~24(Node 24.x). The repo enforces this throughpackage.json#engines. - pnpm:
10.33.x. The repo pinspnpm@10.33.2throughpackageManager; use Corepack so the pinned version is selected automatically. - OS: macOS, Linux, and WSL2 are the primary paths. Windows native is supported; see
docs/windows-troubleshooting.mdfor common setup gotchas. - Optional local agent CLI: Claude Code, Codex, Devin for Terminal, Gemini CLI, OpenCode, Cursor Agent, Qwen, Qoder CLI, GitHub Copilot CLI, etc. If none are installed, use the BYOK API mode from Settings.
Local agent CLI and PATH
The daemon scans your PATH (plus common user toolchain directories). If you install a CLI with npm install -g or Homebrew and Open Design still shows it as not installed, the GUI may be starting with a minimal PATH that does not include your global npm or Homebrew bin directory (common on macOS when the app is not launched from a full login shell). Ensure the executable’s directory is on PATH for the process that runs the daemon, then use Rescan in Settings → Execution mode.
nvm / fnm are optional convenience tools, not required project setup. If you use one, install/select Node 24 before running pnpm:
# nvm
nvm install 24
nvm use 24
# fnm
fnm install 24
fnm use 24
Then enable Corepack and let the repo select pnpm:
corepack enable
corepack pnpm --version # should print 10.33.2
Docker Setup
Run Open Design in a fully containerised environment without installing Node.js or pnpm locally.
Requirements
- Docker Desktop
- Docker Compose v2
Verify Docker is installed correctly:
docker compose version
Start Open Design
From the repository root:
cd deploy
docker compose up -d
Open the app in your browser:
http://localhost:7456
The first startup may take a few seconds while Docker pulls the latest image.
Common Docker Commands
View logs
docker compose logs -f
Restart containers
docker compose restart
Stop containers
docker compose down
Pull the latest image
docker compose pull
docker compose up -d
Remove all local app data
docker compose down -v
Environment Configuration
Create a deploy/.env file to override the default configuration:
# Port exposed on the host
OPEN_DESIGN_PORT=7456
# Container memory limit
OPEN_DESIGN_MEM_LIMIT=384m
# Allowed CORS origins
OPEN_DESIGN_ALLOWED_ORIGINS=https://yourdomain.com
# Docker image tag
OPEN_DESIGN_IMAGE=docker.io/vanjayak/open-design:latest
Persistent Storage
Open Design stores projects and SQLite data inside a Docker volume:
open_design_data
The volume is mounted to:
/app/.od
Data persists across container restarts and image updates.
Inspect the volume:
docker volume inspect open-design_open_design_data
Notes
- Docker mode is ideal for contributors who do not want a local Node.js or pnpm setup.
- The container exposes the production daemon build directly on port
7456. - For development workflows and advanced local setup, see the rest of this Quickstart guide.
One-shot (dev mode)
corepack enable
pnpm install
pnpm tools-dev run web # starts daemon + web in the foreground
# open the web URL printed by tools-dev
For the desktop shell and all managed sidecars in the background:
pnpm tools-dev # starts daemon + web + desktop in the background
On first load, the app detects your installed code-agent CLI (Claude Code / Codex / Devin for Terminal / Gemini / OpenCode / Cursor Agent / Qwen / Qoder CLI), picks it automatically, and defaults to web-prototype skill + Neutral Modern design system. Type a prompt and hit Send. The agent streams into the left pane; the <artifact> tag is parsed out and the HTML renders live on the right. When it finishes, click Save to disk to persist the artifact under ./.od/artifacts/<timestamp>-<slug>/index.html.
The Design system dropdown ships with 71 built-in systems — 2 hand-authored starters (Neutral Modern, Warm Editorial) and 69 product systems imported from awesome-design-md, grouped by category (AI & LLM, Developer Tools, Productivity, Backend, Design Tools, Fintech, E-Commerce, Media, Automotive). Pick one to skin every prototype in that brand's aesthetic, and another set of 57 design skills sourced from awesome-design-skills.
The Skill dropdown groups by mode (Prototype / Deck / Template / Design system) and shows the default skill per mode with a · default suffix. Bundled skills:
- Prototype —
web-prototype(generic),saas-landing,dashboard,pricing-page,docs-page,blog-post,mobile-app. - Deck / PPT —
simple-deck(single-file horizontal swipe) andmagazine-web-ppt(theguizang-pptbundle fromop7418/guizang-ppt-skill— default for deck mode, ships its own assets/template + 4 references). Skills with side files get an automatic "Skill root (absolute)" preamble so the agent can resolveassets/template.htmlandreferences/*.mdagainst the real on-disk path instead of its CWD.
Pair a skill with a design system and a single prompt produces a layout-appropriate prototype or deck in the chosen visual language.
Other scripts
pnpm tools-dev # daemon + web + desktop in the background
pnpm tools-dev start web # daemon + web in the background
pnpm tools-dev run web # daemon + web in the foreground (e2e/dev server)
pnpm tools-dev restart # restart daemon + web + desktop
pnpm tools-dev restart --daemon-port 7457 --web-port 5175
pnpm tools-dev status # inspect managed runtimes
pnpm tools-dev logs # show daemon/web/desktop logs
pnpm tools-dev check # status + recent logs + common diagnostics
pnpm tools-dev stop # stop managed runtimes
pnpm --filter @open-design/daemon build # build apps/daemon/dist/cli.js for `od`
pnpm --filter @open-design/web build # build the web package when needed
pnpm typecheck # workspace typecheck
pnpm tools-dev is the only local lifecycle entry point. Do not use the removed legacy root aliases (pnpm dev, pnpm dev:all, pnpm daemon, pnpm preview, pnpm start).
During local development, tools-dev starts the daemon first, passes its port into apps/web, and apps/web/next.config.ts rewrites /api/*, /artifacts/*, and /frames/* to that daemon port so the App Router app can talk to the sibling Express process without CORS setup.
Media generation / agent dispatcher checks
Image, video, audio, and HyperFrames skills call the local od CLI through environment variables injected by the daemon when it spawns an agent:
OD_BIN— absolute path toapps/daemon/dist/cli.js.OD_DAEMON_URL— the running daemon URL.OD_PROJECT_ID— the active project id.OD_PROJECT_DIR— the active project's file directory.
If media generation fails with OD_BIN: parameter not set, apps/daemon/dist/cli.js missing, or failed to reach daemon at http://127.0.0.1:0, rebuild the daemon CLI and restart the managed runtime:
pnpm --filter @open-design/daemon build
pnpm tools-dev restart --daemon-port 7457 --web-port 5175
ls -la apps/daemon/dist/cli.js
curl -s http://127.0.0.1:7457/api/health
Then open the project from the Open Design app again instead of resuming an old terminal agent session. A daemon-spawned agent should see values like:
echo "OD_BIN=$OD_BIN"
echo "OD_PROJECT_ID=$OD_PROJECT_ID"
echo "OD_PROJECT_DIR=$OD_PROJECT_DIR"
echo "OD_DAEMON_URL=$OD_DAEMON_URL"
ls -la "$OD_BIN"
OD_DAEMON_URL must be a real daemon port such as http://127.0.0.1:7457, not http://127.0.0.1:0. The :0 value is only an internal "pick a free port" launch hint and should not leak into agent sessions.
For the daemon-only production mode, the daemon serves the static Next.js export itself at http://localhost:7456, so no reverse proxy is involved.
If you place nginx in front of the daemon, keep SSE routes unbuffered and uncompressed. A common failure is the browser console showing net::ERR_INCOMPLETE_CHUNKED_ENCODING 200 (OK) after 80-90 seconds because nginx gzip on buffers chunked SSE responses even when the daemon sends X-Accel-Buffering: no.
location /api/ {
proxy_pass http://127.0.0.1:7456;
proxy_buffering off;
gzip off;
proxy_read_timeout 86400s;
proxy_send_timeout 86400s;
proxy_http_version 1.1;
proxy_set_header Connection "";
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
}
Two execution modes
| Mode | Picker value | How a request flows |
|---|---|---|
| Local CLI (default when daemon detects an agent) | "Local CLI" | Frontend → daemon /api/chat → spawn(<agent>, ...) → stdout → SSE → artifact parser → preview |
| API mode (fallback / no CLI) | "Anthropic API" / "OpenAI API" / "Azure OpenAI" / "Google Gemini" | Frontend → daemon /api/proxy/{provider}/stream → provider SSE normalized to delta/end/error → artifact parser → preview |
Both modes feed the same <artifact> parser and the same sandboxed iframe. The only thing that differs is the transport and the system-prompt delivery (local CLIs have no separate system channel, so the composed prompt is folded into the user message).
Prompt composition
For every send, the app builds a system prompt from three layers and sends it to the provider:
BASE_SYSTEM_PROMPT (output contract: wrap in <artifact>, no code fences)
+ active design system body (DESIGN.md — palette/type/layout)
+ active skill body (SKILL.md — workflow and output rules)
Swap the skill or the design system in the top bar and the next send uses the new stack. Bodies are cached in-memory per session so this is a single daemon fetch per pick.
File map
open-design/
├── apps/
│ ├── daemon/ # Node/Express — spawns local agents + serves APIs
│ │ └── src/
│ │ ├── cli.ts # `od` bin entry
│ │ ├── server.ts # /api/* + static serving
│ │ ├── agents.ts # PATH scanner for claude/codex/devin/gemini/opencode/cursor-agent/qwen/qoder/copilot
│ │ ├── skills.ts # SKILL.md loader (frontmatter parser)
│ │ └── design-systems.ts # DESIGN.md loader
│ │ ├── sidecar/ # tools-dev daemon sidecar wrapper
│ │ └── tests/ # daemon package tests
│ ├── web/ # Next.js 16 App Router + React client
│ ├── app/ # App Router entrypoints
│ ├── src/ # React + TypeScript client/runtime modules
│ │ ├── App.tsx # orchestrates mode / skill / DS pickers + send
│ │ ├── providers/ # daemon + BYOK API transports
│ │ ├── prompts/ # system, discovery, directions, deck framework
│ │ ├── artifacts/ # streaming <artifact> parser + manifests
│ │ ├── runtime/ # iframe srcdoc, markdown, export helpers
│ │ └── state/ # localStorage + daemon-backed project state
│ ├── sidecar/ # tools-dev web sidecar wrapper
│ └── next.config.ts # tools-dev rewrites + prod apps/web/out export config
│ └── desktop/ # Electron runtime, launched/inspected by tools-dev
├── packages/
│ ├── contracts/ # shared web/daemon app contracts
│ ├── sidecar-proto/ # Open Design sidecar protocol contract
│ ├── sidecar/ # generic sidecar runtime primitives
│ └── platform/ # generic process/platform primitives
├── tools/dev/ # `pnpm tools-dev` lifecycle and inspect CLI
├── e2e/ # Playwright UI + external integration/Vitest harness
├── skills/ # SKILL.md — drops in from any Claude Code skill repo
│ ├── web-prototype/ # generic single-screen prototype (default for prototype mode)
│ ├── saas-landing/ # marketing page (hero / features / pricing / CTA)
│ ├── dashboard/ # admin / analytics dashboard
│ ├── pricing-page/ # standalone pricing + comparison
│ ├── docs-page/ # 3-column documentation layout
│ ├── blog-post/ # editorial long-form
│ ├── mobile-app/ # phone-frame single screen
│ ├── simple-deck/ # minimal horizontal-swipe deck
│ └── guizang-ppt/ # magazine-web-ppt — bundled deck/PPT default
│ ├── SKILL.md
│ ├── assets/template.html
│ └── references/{themes,layouts,components,checklist}.md
├── design-systems/ # DESIGN.md — 9-section schema (awesome-claude-design)
│ ├── default/ # Neutral Modern (starter)
│ ├── warm-editorial/ # Warm Editorial (starter)
│ ├── README.md # catalog overview
│ └── …129 systems # 2 starters · 70 product systems · 57 design skills
├── scripts/sync-design-systems.ts # re-import from upstream getdesign tarball
├── docs/ # product vision + spec
├── .od/ # runtime data (gitignored, auto-created)
│ ├── app.sqlite # projects / conversations / messages / tabs
│ ├── artifacts/ # one-off "Save to disk" renders
│ └── projects/<id>/ # per-project working dir + agent cwd
├── pnpm-workspace.yaml # apps/* + packages/* + tools/* + e2e
└── package.json # root quality scripts + `od` bin
Troubleshooting
better-sqlite3fails to load / ABI mismatch after a Node.js version change —pnpm installre-runspostinstallautomatically and rebuilds the native addon for the current Node.js. To rebuild manually or verify the fix:pnpm --filter @open-design/daemon rebuild better-sqlite3thenpnpm --filter @open-design/daemon exec node -e "require('better-sqlite3')". Requires build tools:python3,make,g++(orclang++). If you haveignore-scripts=truein your.npmrc, runnode scripts/postinstall.mjsafterpnpm install.- "no agents found on PATH" — install one of:
claude,codex,devin,gemini,opencode,cursor-agent,qwen,qodercli,copilot. Or switch to API mode in Settings and paste a provider key. - Claude Code exits with code 1 — Open Design was able to start
claude, but the spawned non-interactive run failed before producing a response. From the same shell or app environment that starts Open Design, check:
If the smoke test reportsclaude --version claude auth status --text printf 'hello' | claude -p --output-format stream-json --verbose --permission-mode bypassPermissions401,apiKeySource: "none", or another auth error without a custom endpoint, runclaude, use/login, exit Claude, and retry Open Design. If you use multiple Claude profiles, set Settings -> Execution mode -> Claude Code config directory to the profile path such as~/.claude-2. IfANTHROPIC_BASE_URLor a proxy is set, check the endpoint URL, proxy credentials, endpoint auth environment, and model access; remove the custom endpoint only if you want to retry with standard Claude Code auth. On Windows, native PowerShell and WSL use separate Claude installs and credential stores; re-authenticate in the same environment Open Design uses, and check Windows Credential Manager if/logindoes not repair native Windows credentials. - daemon 500 on /api/chat — check the daemon terminal for the stderr tail; usually the CLI rejected its args. Different CLIs take different argv shapes; see
apps/daemon/src/agents.tsbuildArgsif you need to tweak. - media generation says
OD_BINis missing or daemon URL is:0— run the media dispatcher checks above. Do not resume the old CLI session; reopen the project from the Open Design app so the daemon can inject freshOD_*variables. - Codex loads too much plugin context — start Open Design with
OD_CODEX_DISABLE_PLUGINS=1 pnpm tools-devto make daemon-spawned Codex processes run with--disable plugins. - artifact never renders — the model produced text without wrapping in
<artifact>. Confirm the system prompt is going through (check daemon log) and consider switching to a more capable model or a stricter skill.
Mapping back to the vision
This Quickstart is the runnable seed of the spec in docs/. The spec describes where this grows (see docs/roadmap.md). Highlights:
docs/architecture.mddescribes the shipped stack: Next.js 16 App Router in front, local daemon behind it, andapps/web/next.config.tsrewrites in dev to keep the browser talking to the same/apisurface.docs/skills-protocol.mddescribes the fullod:frontmatter (typed inputs, sliders, capability gating). This MVP readsname/description/triggers/od.mode/od.design_system.requiresonly — extendapps/daemon/src/skills.tsto add the rest.docs/agent-adapters.mdforesees richer dispatch (capability detection, streaming tool-calls). Ourapps/daemon/src/agents.tsis a minimal dispatcher — enough to prove the wiring.docs/modes.mdlists four modes: prototype / deck / template / design-system. We ship skills for the first two; the picker already filters bymode.