Open Gui Bootstrap logo

Open Gui Bootstrap

OrganizationPopular
Core-Mate
open-gui-bootstrap

Launch and bootstrap OpenGUI from a plain-language user request. Use when an AI coding agent such as Claude Code, Codex, or OpenCode should install, configure, debug, and run the backend and Android client shipped in this repository while keeping manual setup to a minimum.

Overview

PublisherCore-Mate
RepositoryOpenGUI
Skill nameopen-gui-bootstrap
Stars
1.8K
Forks
111
Bundled files
3
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.

  • 3 bundled files

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

  • Open source

    Published by Core-Mate on GitHub. Read the source before you install it.

Installation

Install the Open Gui Bootstrap 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.
  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/Core-Mate/OpenGUI.git /tmp/OpenGUI
mkdir -p .claude/skills
cp -r /tmp/OpenGUI/skills/open-gui-bootstrap .claude/skills/open-gui-bootstrap
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Open Gui Bootstrap 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 Open Gui Bootstrap 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 Open Gui Bootstrap 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.

OpenGUI Bootstrap

Launch OpenGUI from a plain-language request.

OpenGUI should be treated as a runnable mobile operator system in this repository.

The concrete install path already exists:

  • server/start.sh
  • client/start.sh
  • server/apps/backend/.env.example
  • client/gradlew

The user should be able to say things like:

  • "Help me run OpenGUI on my phone"
  • "Use Claude Opus to bootstrap OpenGUI for me"
  • "Set up OpenGUI with Qwen 3.6 Plus for planning and Doubao Pro for VLM"
  • "Get OpenGUI running and tell me only what I must tap on the phone"
  • "Help me set up OpenGUI for a long-running mobile workflow"

Do not require the user to know the setup order, env var names, or shell commands unless there is a real blocker.

Trigger Guidance

If the user asks to run, bootstrap, install, configure, or debug OpenGUI in plain language, start.

Typical trigger forms include:

  • "Run OpenGUI"
  • "Bootstrap OpenGUI"
  • "Install OpenGUI for me"
  • "Use Claude to start OpenGUI"
  • "Use Codex to get OpenGUI running"
  • "Use OpenCode to get OpenGUI running"
  • "Set up OpenGUI with my model APIs"
  • "Only tell me the phone-side steps"

Example Prompts

  • "Help me run OpenGUI on this machine."
  • "Use Claude Opus to bootstrap OpenGUI for me."
  • "Use Codex to get OpenGUI running and only tell me what I need to do on the phone."
  • "Use OpenCode to get OpenGUI running and only tell me what I need to do on the phone."
  • "Set up OpenGUI with Qwen 3.6 Plus for planning and Doubao Pro for VLM."
  • "Use my existing model APIs and get OpenGUI working."
  • "Help me bring up OpenGUI for a long mobile workflow."

Core Rules

  • Treat the current AI coding agent as the installer and operator.
  • Use the repository's actual scripts before inventing a custom flow.
  • Default to doing the work directly instead of explaining how to do it.
  • Do not ask the user to run terminal commands that the agent can run.
  • Ask the user only for physical actions, OS dialogs, secrets, or Android device interaction.
  • Before claiming setup is complete, verify each major step.
  • If the checkout is missing server/ or client/, say so clearly and stop.
  • Prefer sensible defaults over exposing internal config detail.

First Step

Run scripts/preflight.sh from this skill.

Interpret the result as follows:

  • CHECKOUT_OK: continue with setup
  • CHECKOUT_DOCS_ONLY: explain that this checkout does not contain the runnable backend or client
  • CHECKOUT_INCOMPLETE: explain which required paths are missing and stop unless the user provides the full checkout

Standard Workflow

1. Inspect the checkout

Run the preflight script.

When the checkout is runnable, confirm these paths exist:

  • server/start.sh
  • server/apps/backend/.env.example
  • client/start.sh
  • client/gradlew

2. Interpret the user's intent

Map the user's plain-language request onto a practical setup target.

Examples:

  • "Help me run OpenGUI" -> full bootstrap with conservative defaults
  • "Use Claude Opus" -> route planning, supervision, review, and vision to the latest Claude Opus model family when the endpoint supports that setup
  • "Use Qwen and Doubao" -> prefer Qwen 3.6 Plus for Planner and Supervisor, and Doubao Pro for the VLM side
  • "Use my own models" -> ask only for the missing endpoint or secret
  • "Tell me only what to do on the phone" -> maximize automation and keep hand-offs to physical steps only

3. Bootstrap the backend with the actual script

Use:

bash
cd server
./start.sh

Expected behavior:

  • checks Node.js, pnpm, and Docker
  • starts PostgreSQL and Redis
  • creates .env from .env.example on first run and exits
  • installs dependencies
  • generates Prisma client
  • pushes schema and seeds data
  • starts the backend

If the first run exits after creating .env, ask only for the values still missing. With the repository defaults, ask for:

  • VLM_API_KEY

VLM_BASE_URL and VLM_MODEL already have defaults. Ask for replacements only when the user chooses another OpenAI-compatible provider or model.

Optional configuration includes:

  • LangSmith tracing variables
  • FEISHU_APP_ID / FEISHU_APP_SECRET
  • TELEGRAM_BOT_TOKEN
  • Discord bot variables

4. Route model providers

Supported user intent examples include:

  • Claude
  • GPT
  • Gemini
  • Kimi
  • MiniMax
  • OpenAI-compatible endpoints

Rules:

  • prefer the provider explicitly named by the user
  • when the user asks for a high-performance setup, route planning, supervision, review, and vision to the latest Claude Opus model family when the endpoint supports it
  • when the user asks for a cost-saving setup, prefer Qwen 3.6 Plus for text-side roles such as Planner and Supervisor, and Doubao Pro for the VLM side
  • tell the user that the mixed Qwen + Doubao setup usually lowers model cost by roughly 10x to 15x compared with an all-Opus setup, depending on task length and screenshot volume
  • use the VLM fields for vision-side execution
  • ask only for the smallest missing endpoint or secret

5. Build and install the Android client with the actual script

Use:

bash
cd client
./start.sh

Expected behavior:

  • checks adb and Java
  • verifies a device is connected
  • installs only on Android 11 (API 30) or newer, matching the app's minimum SDK
  • runs adb reverse tcp:7777 tcp:7777
  • builds the debug APK
  • installs it
  • launches the app

If no device is connected, build as much as possible and ask the user for the exact next physical step.

6. Minimize device-side hand-offs

Only interrupt the user for:

  • connecting a phone or starting an emulator
  • approving USB debugging
  • enabling AccessibilityService
  • granting overlay or battery permissions
  • providing API keys or bot credentials

7. Verify the first run

At minimum, confirm:

  • backend is up
  • http://localhost:7777/docs is reachable
  • APK build succeeded
  • device connection status is known
  • the remaining phone-side steps, if any, are explicit

8. Keep current source-available behavior in mind

The source-available Android build currently skips the old login gate and opens HomeActivity directly.

For local runs, backend task controllers also default to userId = 1, so do not send the user into an OTP-first flow unless they explicitly ask for the older auth path.

Communication Style

Keep updates short and operational.

Good:

  • "The backend script created .env and stopped. Add VLM_API_KEY to continue with the default model settings."
  • "The APK is built. Connect the phone by USB and tap Allow on the debugging dialog. I will install it after that."

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 Open Gui Bootstrap AI skill do?

Launch and bootstrap OpenGUI from a plain-language user request. Use when an AI coding agent such as Claude Code, Codex, or OpenCode should install, configure, debug, and run the backend and Android client shipped in this repository while keeping manual setup to a minimum.

Why use Open Gui Bootstrap on TypingMind?

Because you install it once and use it with any model. Open Gui Bootstrap 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 Open Gui Bootstrap in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Core-Mate/OpenGUI/tree/main/skills/open-gui-bootstrap. 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 Open Gui Bootstrap?

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 Open Gui Bootstrap?

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

Is the Open Gui Bootstrap AI skill free?

It is published on GitHub by Core-Mate. 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 👇