Concierge opened in-app support you can run from MCP

Concierge opened a native support chat for iOS, Android, React Native, Expo, Flutter, and a website.

A user opens it from a button, row, or menu. An AI agent answers first from the docs you add, such as FAQs, pricing, and known issues, and the reply is labeled as AI. If someone asks for a person, the thread moves to the top of the inbox and is marked high priority. Reply yourself and the agent pauses on that thread. The SDK makes no network call when the app launches. It registers the first time the chat opens. A client tool, such as Restore purchases, runs only after the user allows it.

iOS uses SwiftUI and needs iOS 16 or later, with no extra dependencies. Android uses Kotlin and Jetpack Compose, with minSdk 24. The React Native and Expo package wraps those native chats, and Expo Go can run a JavaScript fallback before a development build. The Flutter plugin, concierge_sdk, wraps the same chats and needs iOS 16.4 or later. Flutter web is not part of that plugin. The website widget is one script tag, or the @conciergesdk/web package, drawn in a shadow DOM. It loads about 27 KB gzip with the page and does not call the API until someone opens the chat.

You add the app in the dashboard, then paste one setup prompt into a coding agent. The prompt installs the SDK with your app key. You can also run npx @conciergesdk/cli init, which waits for a test message from your phone.

The inbox is a remote MCP server at mcp.conciergesdk.com/mcp, with 51 tools that read and write over Streamable HTTP. Claude Code, Cursor, and Codex can sign in with OAuth. Windsurf, VS Code with GitHub Copilot, Gemini CLI, Zed, and other clients use a staff API key. A reply from the agent shows up in the app live and by push. Triage groups open threads and drafts replies. Nothing sends until you approve it. The site's Claude Code recording lists open threads in a sample app, reads a locked-purchase report, changes the restore code, replies, and flags another thread as high priority.

There is no free plan and no trial. Before you pay, you can sign up, add apps, write help articles, and send test messages from up to 3 of your own devices. AI replies and messages from real users start when you subscribe.

Pro is one price for unlimited apps and seats, with 1,000 AI replies a month. The first 50 organizations to pay get $19 a month, or $190 a year, and that price stays while the subscription stays active. Canceling ends the lock. After those 50, Pro is $49 a month, or $400 a year. Prices are in US dollars and exclude tax. When the 1,000 replies are used, AI replies pause until the next billing month. Concierge does not add an overage charge unless you agree to one. Scale is priced on request. The refund desk and billing heads-ups stay off until you turn them on. Two-way translation, Slack replies, and in-app analytics are marked Soon. A packaged server for self-hosting is not part of the product.