mobile-app-builder-ai（移动应用程序生成器 ai） Skills下载安装命令和教程

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手，先确认命令和权限，再让它执行。

请帮我安装这个 Agent Skill：mobile-app-builder-ai（移动应用程序生成器 ai）
来源仓库：https://github.com/versacexcodes/mobile-app-builder-ai
安装命令：
openclaw skills install mobile-app-builder-ai
安装前请先检查当前环境是否支持对应 CLI，并向我确认将要执行的命令、安装目录、联网范围和文件读写权限；确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill；本站只展示命令，不托管安装包，也不自动执行。

ClawHubOpenClaw

openclaw skills install mobile-app-builder-ai

简介

根据文本描述构建全栈网络和移动应用程序。创建项目、规划功能并启动快速入门构建以进行后台执行。

SKILL.md

name: launchpulse
version: 1.1.1
description: Build full-stack web and mobile apps from a text description. Creates projects, plans features, and starts Quick Start builds for background execution.
homepage: https://launchpulse.ai
license: MIT
user-invocable: true
metadata: {"openclaw":{"homepage":"https://launchpulse.ai","requires":{"bins":["node"]},"author":"LaunchPulse","categories":["development","deployment"]}}

LaunchPulse (Web + Mobile App Generator)

Build production-ready web apps and Expo mobile apps from a text description. LaunchPulse plans features, starts Quick Start in the background, and returns project/session context immediately.

What it does:

Creates a new project (web or Expo mobile)
Plans MVP features using AI and auto-approves them
Starts Quick Start (single-pass background build)
Returns immediately with project id/session id so progress can be monitored
Iterates on existing projects with change requests
Supports production ops: deployments, app store publish, domains, database, storage, env/payment setup

If feature planning is temporarily unavailable, the skill falls back to a single MVP feature so Quick Start never stalls.

Authentication

The first time you run the skill, it automatically starts a device-login flow:

OpenClaw prints a sign-in link
You open it, sign in with Google on launchpulse.ai, and confirm
OpenClaw receives a personal access token (PAT) and stores it locally

Token storage path: ${OPENCLAW_STATE_DIR:-~/.openclaw}/launchpulse/auth.json

Advanced alternatives:

LAUNCHPULSE_PAT / --pat <token>: provide a PAT directly
LAUNCHPULSE_API_KEY / --api-key <token>: alias for PAT-style API key auth
LAUNCHPULSE_ACCESS_TOKEN / --access-token <jwt>: provide a Supabase JWT (advanced)

Billing

LaunchPulse uses a token-based billing system. Every AI operation (feature planning, building, iterating) consumes tokens.

Plans:

Plan	Price	Tokens/month	Best for
Free	$0	10,000 (~2-5 features)	Trying it out
Starter	$19/mo	200,000 (~50-100 features)	Individual builders
Builder	$49/mo	500,000 (~125-250 features)	Serious app builders

Token addon packs (one-time, never expire):

50K tokens: $5
100K tokens: $9
500K tokens: $39 (22% off)
1M tokens: $69 (31% off)

What happens when you run out:

Before Quick Start/iterate starts, the skill checks your balance and blocks with a clear message if exhausted
If tokens run out during a build, the backend will stop progress and report billing context in status APIs
Run /launchpulse status to check your balance anytime
Run /launchpulse upgrade to purchase more tokens or upgrade your plan

How to use

Login (optional, happens automatically on first use):

/launchpulse login

Logout (clears stored token):

/launchpulse logout

Check balance:

/launchpulse status

Upgrade plan or buy tokens:

/launchpulse upgrade (shows all options)
/launchpulse upgrade --tier STARTER (subscribe to Starter)
/launchpulse upgrade --tier BUILDER --billing-period annual (annual Builder)
/launchpulse upgrade --tokens 0 (buy 50K token pack)

Quick Start a web app:

/launchpulse web "Create a landing page for a dog walking business"

Quick Start a mobile app (Expo):

/launchpulse mobile "Create an Expo app for habit tracking"

Iterate on an existing project:

/launchpulse iterate <projectId> "Add a contact form to the homepage"

Production operations:

/launchpulse projects
/launchpulse project-status <projectId>
/launchpulse deploy <projectId> --target cloud-run --wait
/launchpulse deploy-status <projectId> <deploymentId>
/launchpulse store-publish <projectId> --payload-file ./store-publish.json --wait
/launchpulse store-status <projectId> [--publish-id <id>]
/launchpulse store-2fa <projectId> <publishId> <code>
/launchpulse domains search "mybrand"
/launchpulse domains check mybrand.com
/launchpulse domains add <projectId> mybrand.com
/launchpulse domains provision <projectId> mybrand.com my-fly-app
/launchpulse db info <projectId>
/launchpulse db table <projectId> users
/launchpulse db query <projectId> "select * from users limit 10"
/launchpulse storage init
/launchpulse storage upload <projectId> --payload-file ./upload.json
/launchpulse env-files list <projectId>
/launchpulse env-files save <projectId> --file-path vitereact/.env --vars-file ./vars.json
/launchpulse payments inject-env <projectId> --project-type vitereact
/launchpulse payments setup <projectId> --project-type expo --stripe-publishable-key pk_test_... --revenuecat-ios-key appl_... --revenuecat-secret-key sk_...

If the user says "iterate it" without an id, use the most recent projectId from the current chat/session.

What the assistant should do internally

Run the script from the OpenClaw workspace root:

cd ~/.openclaw/workspace
node skills/launchpulse/scripts/launchpulse.cjs login
node skills/launchpulse/scripts/launchpulse.cjs status
node skills/launchpulse/scripts/launchpulse.cjs upgrade --tier STARTER
node skills/launchpulse/scripts/launchpulse.cjs web "<description>"
node skills/launchpulse/scripts/launchpulse.cjs mobile "<description>"
node skills/launchpulse/scripts/launchpulse.cjs iterate <projectId> "<change request>"
node skills/launchpulse/scripts/launchpulse.cjs projects
node skills/launchpulse/scripts/launchpulse.cjs project-status <projectId>
node skills/launchpulse/scripts/launchpulse.cjs deploy <projectId> --target cloud-run --wait
node skills/launchpulse/scripts/launchpulse.cjs store-publish <projectId> --payload-file ./store-publish.json --wait
node skills/launchpulse/scripts/launchpulse.cjs domains search "<query>"
node skills/launchpulse/scripts/launchpulse.cjs db info <projectId>
node skills/launchpulse/scripts/launchpulse.cjs payments setup <projectId> --project-type vitereact --vars-file ./vars.json

Useful options:

--api-base <url>: override API base (for local/dev backend)
--name <slug>: preferred project id
--pat <token> / --access-token <jwt>: advanced auth overrides
--no-plan: skip AI planner and use MVP fallback feature directly
--timeout-min <n>: timeout in minutes for long-running commands (iterate/deploy/store wait)
--chat-id <id>: (iterate) use a specific project chat
--provider <id> / --model <id>: (iterate) override AI provider/model
--target <cloud-run|fly>: (deploy) deployment target
--wait: poll until terminal status (deploy/store publish)
--payload-file <path>: JSON payload for complex commands
--project-type <vitereact|expo>: payment/env injection mode
--vars-file <path> / --file-path <path>: env-files save/payment setup

User updates (required)

When running this skill for users in Telegram/OpenClaw, keep updates short and frequent:

Send a milestone update after each major step: auth, project creation, planning/fallback, quick-start kickoff.
When script returns status: "quick_start_started", always send an explicit completion update immediately with projectId, sessionId, and guidance to monitor via project-status.
If stuck or degraded (planner failure, quick-start start failure), say exactly what happened and what is still usable.
If status: "token_limit_exceeded", tell the user their tokens are used up and share the upgrade URL from the output.

Output

The script prints a final JSON object to stdout including:

ok: boolean success flag
status: operation-specific status (quick_start_started, success, failed, token_limit_exceeded, deploy/store/domain/db/payment states)
projectId, mode, sessionId
preview: URLs (frontendUrl, backendUrl, expoGoUrl)
features: counts snapshot (total, selected, completed)
billing: (when applicable) tier, upgradeUrl, buyTokensUrl

Exit codes: 0=success, 1=error, 2=paused, 3=timeout, 4=failed, 5=billing/token limit

Backend selection

Default API: https://api.launchpulse.ai/api

Override for local dev:

LAUNCHPULSE_API_BASE_URL=http://localhost:667/api
or --api-base http://localhost:667/api

Changelog

1.1.1

Switched web and mobile flow to Quick Start only
Removed auto-mode loop polling/continue behavior from create flow
web/mobile now return immediately with status: "quick_start_started"

1.1.0

Added production commands: projects, project-status, deploy/deploy-status
Added store publish support (start/status/2FA/wait)
Added domains command suite (search/check/map/provision/status/dns/checkout/register/verification)
Added database command suite (info/table/query)
Added storage command suite (init/upload/save-assets)
Added env-files and payment setup commands (inject env + Stripe/RevenueCat env save)
Hardened upgrade failure exit behavior (non-zero on checkout fallback failure)

1.0.0

Initial ClawHub release
Device auth flow (Google sign-in)
Web and mobile (Expo) project creation
AI-powered feature planning with fallback
Auto-build with live polling
Project iteration support
Token balance checking and billing integration
status command to check plan and usage
upgrade command with direct Stripe checkout links

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明，方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用；具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

87.3%

按下载量换算5,956

安全审计

VirusTotal

可疑

ClawScan

可疑

Static analysis

未展示

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置，应进入高风险复核队列，默认不自动发布。

安装前确认

本站仅展示第三方公开信息，不托管安装包，不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果，不能写成本站确认安全。当前只有一个来源，正式发布前建议补源仓库或其他目录站核验。

来源信息

GitHub 仓库

github.com

/versacexcodes/mobile-app-builder-ai

SKILL.md 来源

clawhub.ai

/versacexcodes/mobile-app-builder-ai

继续浏览同类 Skills

更多研究检索 Skills（73,675）返回 Skills 目录

上一个 Skill

mobile-android-design（手机安卓设计）

用于辅助界面设计、视觉规范、排版、配色、布局和交互体验优化。它适合让 Agent 根据产品场景整理页面结构、生成 UI 方案、检查视觉一致性或改进组件层级。使用时需要结合现有品牌、设计系统和用户任务，不应只堆装饰元素；涉及真实页面改动时，应通过截图或浏览器预览检查文本溢出、对齐和响应式表现。

下一个 Skill

mobile-app-debugging（移动应用程序调试）

mobile-app-debugging 用于查找、检索和筛选相关信息，适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态，以及是否会触发联网、命令执行或文件读写。