[n8n] Create a mobile app for accessing workflows

I've been running n8n workflows for my business and kept hitting the same problem - accessing them from mobile meant either opening a browser or building a Telegram bot. Both felt like workarounds, not solutions. So I built **ShellX** \- a dedicated mobile app that connects to n8n/Make/Zapier webhooks. **Streaming responses** \- Used SSE so AI responses appear word-by-word instead of dumping a full block of text. Makes it feel like actual chat. **Context switching** \- Added "labels" so one webhook can route to different workflows. Tap \[Home\] to control my house, \[Work\] to query company data, \[Personal\] for my AI assistant. No bot commands, just seamless switching. **Native features** \- Built-in QR scanner, voice notes, push notifications. Stuff that makes sense on mobile but is hard to hack into Telegram. **Privacy approach** \- The app doesn't see your data. It's just a UI layer that talks directly to your webhooks. Everything stays on your infrastructure. **What I'm curious about:** * What workflows would you want to run from mobile? * Is the Telegram workaround actually a problem for others or just me? * Any features you'd want to see? * Live on iOS/Android app stores (free) https://preview.redd.it/tv71umvcigmg1.png?width=768&format=png&auto=webp&s=5e9f5151532d35a88edf73e954373d3067cb1d3e Happy to answer technical questions about the implementation. [shellx.app](http://shellx.app) { "nodes": [ { "parameters": { "promptType": "define", "text": "={{ $items('Webhook')[0].json.body.chatInput }}", "options": { "systemMessage": "You are my assistant.\nAnswer the question directly and only ask a follow-up question if it is strictly necessary.", "enableStreaming": true } }, "type": "@n8n/n8n-nodes-langchain.agent", "typeVersion": 3.1, "position": [-32, -16], "id": "e4503e0f-0815-46ce-8b2e-3b26547cfeb9", "name": "AI Agent Chat", "onError": "continueRegularOutput" }, { "parameters": { "httpMethod": "POST", "path": "shellx-groq-chat", "responseMode": "streaming", "options": {} }, "type": "n8n-nodes-base.webhook", "typeVersion": 2.1, "position": [-240, -16], "id": "ce7a6c2a-bdc9-49c9-99dd-53010f9bee50", "name": "Webhook", "webhookId": "shellx-groq-chat" }, { "parameters": { "sessionIdType": "customKey", "sessionKey": "={{ $items('Webhook')[0].json.body.sessionId }}", "contextWindowLength": {} }, "type": "@n8n/n8n-nodes-langchain.memoryBufferWindow", "typeVersion": 1.3, "position": [64, 160], "id": "d0182671-d9e9-4142-9863-c9b681aff93f", "name": "Simple Memory" }, { "parameters": { "model": "openai/gpt-oss-120b", "options": {} }, "type": "@n8n/n8n-nodes-langchain.lmChatGroq", "typeVersion": 1, "position": [-176, 192], "id": "b221af9c-0245-4e83-846b-20679efeee99", "name": "Groq Chat Model" } ], "connections": { "Webhook": { "main": [ [ { "node": "AI Agent Chat", "type": "main", "index": 0 } ] ] }, "Simple Memory": { "ai_memory": [ [ { "node": "AI Agent Chat", "type": "ai_memory", "index": 0 } ] ] }, "Groq Chat Model": { "ai_languageModel": [ [ { "node": "AI Agent Chat", "type": "ai_languageModel", "index": 0 } ] ] } }, "meta": { "templateCredsSetupCompleted": false } }