Now just OpenClaw gateway with the built-in dashboard.

Dockerfile

@@ -1,10 +1,9 @@
 FROM node:22-slim
 
-# Install system dependencies including git and netcat
+# Install system dependencies
 RUN apt-get update && apt-get install -y \
     curl \
     git \
-    netcat-openbsd \
     python3 \
     build-essential \
     && rm -rf /var/lib/apt/lists/*
@@ -12,18 +11,8 @@ RUN apt-get update && apt-get install -y \
 # Install OpenClaw globally
 RUN npm install -g openclaw
 
-# Create app directory
-WORKDIR /app
-
-# Copy package.json and install dependencies
-COPY package.json /app/package.json
-RUN npm install
-
-# Copy application files
-COPY public /app/public
-COPY server.js /app/server.js
-COPY start.sh /app/start.sh
-RUN chmod +x /app/start.sh
+# Create workspace
+RUN mkdir -p /root/.openclaw/workspace
 
 # Expose port
 EXPOSE 7860
@@ -33,5 +22,10 @@ ENV BASE_URL=""
 ENV API_KEY=""
 ENV MODEL_ID="Multiple-models"
 ENV PROVIDER_NAME="custom"
+ENV GATEWAY_TOKEN="hf-space-token"
+
+# Copy start script
+COPY start.sh /app/start.sh
+RUN chmod +x /app/start.sh
 
 CMD ["/app/start.sh"]

README.md

@@ -1,5 +1,5 @@
 ---
-title: OpenClaw Web Chat
+title: OpenClaw Dashboard
 emoji: 🦞
 colorFrom: blue
 colorTo: purple
@@ -7,106 +7,27 @@ sdk: docker
 pinned: false
 ---
 
-# OpenClaw Web Chat
+# OpenClaw Dashboard
 
-Deploy your AI assistant to Hugging Face Spaces with a web-based chat interface.
+Access your AI assistant through the OpenClaw dashboard.
 
-## Why Web UI?
+## Setup
 
-Hugging Face Spaces free tier blocks outbound HTTPS, which prevents Telegram/Discord integrations. This setup uses a web UI that only requires inbound HTTP (which HF allows).
+1. Set these secrets in your Space:
+   - `BASE_URL` - Your API URL (e.g., `https://your-9router.hf.space/v1`)
+   - `API_KEY` - Your API key
 
-## Setup Instructions
+2. Open your Space URL
 
-### 1. Create Hugging Face Space
+3. When prompted, enter the gateway token: `hf-space-token`
+   (Or set your own via `GATEWAY_TOKEN` secret)
 
-1. Go to [huggingface.co/spaces](https://huggingface.co/spaces)
-2. Click "Create new Space"
-3. Select:
-   - **Space name**: Your choice (e.g., `my-openclaw-chat`)
-   - **License**: Apache-2.0
-   - **Space SDK**: Docker
-   - **Space hardware**: CPU (free tier)
+## Usage
 
-### 2. Upload Files
-
-Upload these files to your Space:
-- `Dockerfile`
-- `start.sh`
-- `web-server.js`
-- `public/index.html`
-- `README.md` (this file)
-
-### 3. Set Secrets
-
-In your Space settings, add these secrets:
-
-| Secret Name | Value | Required |
-|-------------|-------|----------|
-| `BASE_URL` | Your API URL (e.g., `https://your-9router.hf.space/v1`) | **Yes** |
-| `API_KEY` | Your API key | **Yes** |
-| `MODEL_ID` | Model name (default: `Multiple-models`) | No |
-| `PROVIDER_NAME` | Provider name (default: `custom`) | No |
-
-**Example for 9router on Hugging Face Spaces:**
-```
-BASE_URL=https://your-9router-space.hf.space/v1
-API_KEY=your-9router-api-key
-MODEL_ID=Multiple-models
-PROVIDER_NAME=9router
-```
-
-### 4. Access Your Chat
-
-1. Wait for the Space to build and start (check logs)
-2. Open your Space URL in a browser
-3. Start chatting with your AI assistant!
-
-### 5. Keep It Alive (Optional)
-
-Sign up at [UptimeRobot](https://uptimerobot.com) and monitor your Space URL to prevent sleeping.
-
-## Architecture
-
-```
-Browser → HF Space (port 7860) → Web Server
-                     ↓
-            WebSocket Proxy
-                     ↓
-         OpenClaw Gateway (port 18789)
-                     ↓
-              AI Model API
-```
-
-## Features
-
-- ✅ Clean web chat interface
-- ✅ Works on HF Spaces free tier
-- ✅ Real-time responses
-- ✅ Mobile-friendly design
-
-## Troubleshooting
-
-**"Connection error" in browser?**
-- Check Space logs for errors
-- Verify `BASE_URL` and `API_KEY` are set correctly
-- Make sure your API server is accessible from HF Spaces
-
-**Blank page?**
-- Check browser console for JavaScript errors
-- Verify all files were uploaded correctly
-
-**No response from AI?**
-- Check if API is reachable from HF Spaces
-- Verify API key is correct
-- Check Space logs for API errors
+The dashboard provides a full chat interface with your AI assistant (Zero).
 
 ## Files
 
-- `Dockerfile` - Container setup with Node.js + OpenClaw
-- `start.sh` - Startup script (runs gateway + web server)
-- `web-server.js` - Simple HTTP + WebSocket proxy server
-- `public/index.html` - Web chat UI
-
-## Credits
-
-Powered by [OpenClaw](https://github.com/openclaw/openclaw)
+- `Dockerfile` - Container setup
+- `start.sh` - Startup script
+- `README.md` - This file

start.sh

@@ -17,7 +17,10 @@ BASE_URL=${BASE_URL:-"http://127.0.0.1:20128/v1"}
 MODEL_ID=${MODEL_ID:-"Multiple-models"}
 PROVIDER_NAME=${PROVIDER_NAME:-"custom"}
 
-# Create OpenClaw config - bind to loopback only (no pairing needed for local)
+# Token for gateway auth
+GATEWAY_TOKEN=${GATEWAY_TOKEN:-"hf-space-token"}
+
+# Create OpenClaw config
 cat > /root/.openclaw/openclaw.json << EOF
 {
   "meta": {
@@ -54,10 +57,14 @@ cat > /root/.openclaw/openclaw.json << EOF
   },
   "session": { "dmScope": "per-channel-peer" },
   "gateway": {
-    "port": 18789,
+    "port": 7860,
     "mode": "local",
-    "bind": "loopback",
-    "auth": { "mode": "none" },
+    "bind": "lan",
+    "auth": {
+      "mode": "token",
+      "token": "${GATEWAY_TOKEN}"
+    },
+    "trustedProxies": ["0.0.0.0/0"],
     "tailscale": { "mode": "off" }
   },
   "skills": { "install": { "nodeManager": "npm" } },
@@ -65,36 +72,15 @@ cat > /root/.openclaw/openclaw.json << EOF
 }
 EOF
 
-echo "✅ Config created (loopback mode - no pairing)"
+echo "✅ Config created"
 echo "📡 API: ${BASE_URL}"
 echo "🤖 Model: ${MODEL_ID}"
+echo "🔑 Gateway Token: ${GATEWAY_TOKEN}"
+echo ""
+echo "🌐 Starting OpenClaw Gateway..."
+echo "📱 Open your Space URL to access the dashboard"
+echo "🔑 Use token: ${GATEWAY_TOKEN}"
+echo ""
 
-# Start OpenClaw gateway in background
-echo "🚀 Starting OpenClaw Gateway..."
-openclaw gateway > /tmp/gateway.log 2>&1 &
-GATEWAY_PID=$!
-
-# Wait for gateway to be ready
-echo "⏳ Waiting for gateway to start..."
-for i in {1..60}; do
-    if nc -z 127.0.0.1 18789 2>/dev/null; then
-        echo "✅ Gateway is ready!"
-        break
-    fi
-    if [ $i -eq 60 ]; then
-        echo "❌ Gateway failed to start after 60 seconds"
-        echo "📋 Gateway logs:"
-        cat /tmp/gateway.log
-        exit 1
-    fi
-    sleep 1
-done
-
-# Show gateway status
-echo "📋 Gateway startup logs:"
-head -15 /tmp/gateway.log
-
-# Start web proxy server
-echo "🌐 Starting Web Proxy..."
-cd /app
-exec node server.js
+# Start OpenClaw gateway directly
+exec openclaw gateway