Control Ui logo

Control Ui

OrganizationPopular
openclaw
control-ui

Operate and troubleshoot the OpenClaw Control UI: navigate connected clients, organize sessions, build session dashboards, and handle direct or Tailscale-hosted Gateways.

Overview

Publisheropenclaw
Repositoryopenclaw
Skill namecontrol-ui
Stars
391K
Forks
82.2K
Bundled files
2
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by openclaw on GitHub. Read the source before you install it.

Installation

Install the Control Ui AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/openclaw/openclaw/tree/main/skills/control-ui
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/openclaw/openclaw.git /tmp/openclaw
mkdir -p .claude/skills
cp -r /tmp/openclaw/skills/control-ui .claude/skills/control-ui
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Control Ui in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Control Ui on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Control Ui is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Control UI

Use OpenClaw's typed UI tools for state and layout. Use browser automation only to inspect or interact with rendered pixels.

Mental model

  • A sidebar item is a session, not a dashboard page.
  • Each session owns one board. The board contains tabs and widgets.
  • Pinning a session keeps it in the sidebar. Sessions with a stored board appear in the /dashboards gallery.
  • screen changes connected Control UI layout and navigation. It does not read the page or take screenshots.
  • sessions_list finds sessions. sessions renames, groups, pins, or archives them. It does not edit board content.
  • dashboard reads and arranges the current session's board and registered plugin widgets. show_widget authors or updates custom HTML/SVG widgets.

Do not replace these operations with shell calls or raw Gateway RPC when the typed tool exists.

Start safely

  1. Identify the target session and whether the Control UI is local/direct or published through Tailscale or another HTTPS proxy.
    • If the user wants a dedicated sidebar dashboard, choose or create its visible session before authoring any widget. A tab is only a partition inside one session's board; it never creates a sidebar row.
    • If the current turn is running in a temporary or parent session, hand the dashboard task to the intended visible session first. Board widgets are keyed by session and cannot be moved to another session afterward.
  2. Read current state before changing it:
    • use sessions_list to resolve the session;
    • use dashboard with action: "read" for the current session;
    • inspect existing tabs, stable widget names, owners, sizes, and panel layout.
  3. If the task targets another session's board, move the work into that session. Dashboard tools intentionally operate on the current session.

Read hosting.md before opening or repairing a remote Control UI. Read dashboards.md before creating or restructuring a board. Use the visualize skill for widget composition, libraries, interactive code explanations, and dashboard content. Keep session and board organization here.

Navigate and arrange the UI

Use screen for deterministic client commands:

  • navigate to open a session by sessionKey;
  • sidebar_show / sidebar_hide for the session sidebar;
  • split_right / split_down, focus, and close_pane for panes;
  • terminal_show / terminal_hide and browser_show / browser_hide for docked panels.

screen broadcasts to every connected Control UI that advertises UI commands; it cannot select one browser tab. Confirm the blast radius when several clients may be open. If it reports no capable client, ask the user to open the Control UI and retry.

Use the in-app browser or an available browser-control tool when the task needs DOM inspection, clicking, typing, or screenshots. Reuse the existing signed-in Control UI tab when possible.

Build a session dashboard

  1. Call dashboard read. Reuse suitable tabs and widget names.
  2. Create or rename tabs with tab_create / tab_update; use short lowercase slug IDs.
  3. Add content with the correct owner:
    • native text, metrics, tables, charts, and links: show_widget with pin: true and report;
    • self-contained custom HTML/SVG or registered-source content: show_widget with pin: true;
    • trusted plugin widgets: dashboard widget_put with an advertised pluginKind. Do not embed Grafana or another external application in an <iframe> inside show_widget; widget sandboxes reject child-frame URLs and navigation. Use an ordinary user-clicked link to open the application, fetch an exact HTTPS API through declared capabilities.netOrigins, use a Gateway data binding, or install/use a trusted plugin widget instead.
  4. Use a stable name when calling show_widget. Reusing the same name with new widget_code updates the widget in place.
  5. Arrange with widget_move, widget_resize, and tab reordering. Prefer size presets and board order over pixel placement.
  6. Pin the current session with sessions patch when it should stay prominent.
  7. Call dashboard focus_tab to show the intended tab in the dashboard side panel. If no Control UI is connected, the command returns unavailable; have the user open the session and retry.
  8. Choose the presentation after focusing the tab. set_presentation with presentation: "expanded" expands the dashboard; "split" shows it beside chat in the current panel layout. The human can use Expand side panel for a full-width dashboard, Collapse to bring chat back, or close the panel for chat alone.

Never create a fake top-level page for a dashboard. For a dedicated dashboard, use a dedicated session, pin that session, and build its board from inside it.

Update without breaking the board

  • Read first and mutate the smallest unit.
  • Keep stable tab IDs and widget names unless replacement is intentional.
  • Do not change a widget's content owner in place. Remove it, then recreate it with the new owner or registered kind.
  • Capability grants are bound to exact widget bytes and revision. Changed content may require approval again.
  • Resetting a conversation preserves its board. Deleting the session deletes the board.

Verify the result

Use two layers of proof:

  1. State: dashboard read shows the expected tabs, widgets, names, owners, order, and sizes; sessions_list shows the expected session metadata.
  2. Rendered UI: open the actual Control UI, confirm the correct session and dashboard panel, then inspect or screenshot the widget frame. Exercise any important controls.

An HTTP 200 for the shell or widget route is transport proof, not visual proof. Do not claim the dashboard works until the sandboxed frame renders. For widgets that fetch live data, exercise that fetch from the rendered frame. A host-side curl does not prove the browser received the required capability grant or accepted the endpoint's CORS and mixed-content policy. If no browser-control or connected-client inspection is available, report the dashboard as published but visually unverified instead of claiming success.

Recovery order

When a widget is blank or stale:

  1. confirm the expected session, tab, widget name, and revision with dashboard read;
  2. classify the hosting path using hosting.md;
  3. verify the Control UI origin, the separate widget sandbox origin, and the protocol used on every reverse-proxy hop;
  4. if widget JavaScript reports Failed to fetch, verify its exact HTTPS origin is declared and granted in capabilities.netOrigins, then check browser CORS and mixed-content errors;
  5. after a Gateway restart or routing change, republish/update the widget so the browser receives a fresh ticketed frame URL;
  6. reload the Control UI and verify the rendered frame, not only route status.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Control Ui AI skill do?

Operate and troubleshoot the OpenClaw Control UI: navigate connected clients, organize sessions, build session dashboards, and handle direct or Tailscale-hosted Gateways.

Why use Control Ui on TypingMind?

Because you install it once and use it with any model. Control Ui is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Control Ui in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/openclaw/openclaw/tree/main/skills/control-ui. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Control Ui?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Control Ui?

As many as you like. As long as a model supports skills, you can use Control Ui with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Control Ui AI skill free?

It is published on GitHub by openclaw. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇