refactor: Improve the configurations

CLAUDE.md

@@ -10,7 +10,7 @@ browser-agent is an A2A (Agent-to-Agent) server implementing the [A2A Protocol](
 
 ### ADL-Generated Structure
 
-The codebase is generated using ADL CLI 0.23.8 and follows a strict generation pattern:
+The codebase is generated using ADL CLI 0.23.9 and follows a strict generation pattern:
 - **Generated Files**: Marked with `DO NOT EDIT` headers - manual changes will be overwritten
 - **Configuration Source**: `agent.yaml` - defines agent capabilities, skills, and metadata
 - **Server Implementation**: Built on the ADK (Agent Development Kit) framework from `github.com/inference-gateway/adk`
@@ -117,7 +117,7 @@ Activate with: `flox activate` (if Flox is installed)
 
 - **Generated Files**: Never manually edit files with "DO NOT EDIT" headers
 - **Configuration Changes**: Always modify `agent.yaml` and regenerate
-- **ADL Version**: Ensure ADL CLI 0.23.8 or compatible version for regeneration
+- **ADL Version**: Ensure ADL CLI 0.23.9 or compatible version for regeneration
 - **Port Configuration**: Default 8080, configurable via `A2A_PORT` or `A2A_SERVER_PORT`
 
 ## Debugging Tips

README.md

@@ -15,12 +15,18 @@ A production-ready [Agent-to-Agent (A2A)](https://github.com/inference-gateway/a
 ## Quick Start
 
 ```bash
-# Run the agent
+# Run the agent locally
 go run .
 
-# Or with Docker
+# Or with Docker (Chromium only - smallest image)
 docker build -t browser-agent .
 docker run -p 8080:8080 browser-agent
+
+# Build with specific browser engine
+docker build --build-arg BROWSER_ENGINE=firefox -t browser-agent:firefox .
+
+# Run with Xvfb enabled (for extensions or specific rendering features)
+docker run -p 8080:8080 -e BROWSER_XVFB_ENABLED=true browser-agent
 ```
 
 ## Features
@@ -62,16 +68,21 @@ The following custom configuration variables are available:
 |----------|----------|-------------|---------|
 | **Browser** | `BROWSER_ARGS` | Args configuration | `[--disable-blink-features=AutomationControlled --disable-features=VizDisplayCompositor --no-first-run --disable-default-apps --disable-extensions --disable-plugins --disable-sync --disable-translate --hide-scrollbars --mute-audio --no-zygote --disable-background-timer-throttling --disable-backgrounding-occluded-windows --disable-renderer-backgrounding --disable-ipc-flooding-protection]` |
 | **Browser** | `BROWSER_DATA_DIR` | Data_dir configuration | `/tmp/playwright/artifacts` |
+| **Browser** | `BROWSER_ENGINE` | Engine configuration | `chromium` |
 | **Browser** | `BROWSER_HEADER_ACCEPT` | Header_accept configuration | `text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7` |
 | **Browser** | `BROWSER_HEADER_ACCEPT_ENCODING` | Header_accept_encoding configuration | `gzip, deflate, br` |
 | **Browser** | `BROWSER_HEADER_ACCEPT_LANGUAGE` | Header_accept_language configuration | `en-US,en;q=0.9` |
 | **Browser** | `BROWSER_HEADER_CONNECTION` | Header_connection configuration | `keep-alive` |
 | **Browser** | `BROWSER_HEADER_DNT` | Header_dnt configuration | `1` |
 | **Browser** | `BROWSER_HEADER_UPGRADE_INSECURE_REQUESTS` | Header_upgrade_insecure_requests configuration | `1` |
 | **Browser** | `BROWSER_HEADLESS` | Headless configuration | `true` |
+| **Browser** | `BROWSER_STEALTH_MODE` | Stealth_mode configuration | `false` |
 | **Browser** | `BROWSER_USER_AGENT` | User_agent configuration | `Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36` |
 | **Browser** | `BROWSER_VIEWPORT_HEIGHT` | Viewport_height configuration | `1080` |
 | **Browser** | `BROWSER_VIEWPORT_WIDTH` | Viewport_width configuration | `1920` |
+| **Browser** | `BROWSER_XVFB_DISPLAY` | Xvfb_display configuration | `:99` |
+| **Browser** | `BROWSER_XVFB_ENABLED` | Xvfb_enabled configuration | `false` |
+| **Browser** | `BROWSER_XVFB_SCREEN_RESOLUTION` | Xvfb_screen_resolution configuration | `1920x1080x24` |
 
 | Category | Variable | Description | Default |
 |----------|----------|-------------|---------|
@@ -158,10 +169,10 @@ docker run --rm -it --network host ghcr.io/inference-gateway/a2a-debugger:latest
 
 ### Docker
 
-The Docker image can be built with custom version information using build arguments:
+The Docker image can be built with custom version information and browser selection using build arguments:
 
 ```bash
-# Build with default values from ADL
+# Build with default values from ADL (Chromium only)
 docker build -t browser-agent .
 
 # Build with custom version information
@@ -170,15 +181,42 @@ docker build \
   --build-arg AGENT_NAME="My Custom Agent" \
   --build-arg AGENT_DESCRIPTION="Custom agent description" \
   -t browser-agent:1.2.3 .
+
+# Build with specific browser engine
+docker build --build-arg BROWSER_ENGINE=firefox -t browser-agent:firefox .
+
+# Build with all browsers (larger image)
+docker build --build-arg BROWSER_ENGINE=all -t browser-agent:all .
 ```
 
 **Available Build Arguments:**
 - `VERSION` - Agent version (default: `0.4.1`)
 - `AGENT_NAME` - Agent name (default: `browser-agent`)
 - `AGENT_DESCRIPTION` - Agent description (default: `AI agent for browser automation and web testing using Playwright`)
+- `BROWSER_ENGINE` - Browser to install (`chromium`, `firefox`, `webkit`, or `all`) (default: `chromium`)
 
 These values are embedded into the binary at build time using linker flags, making them accessible at runtime without requiring environment variables.
 
+#### Xvfb Configuration
+
+By default, the browser runs in native headless mode. For cases requiring a virtual display (e.g., extensions, specific rendering features), you can enable Xvfb:
+
+```bash
+# Run with Xvfb enabled
+docker run -p 8080:8080 \
+  -e BROWSER_XVFB_ENABLED=true \
+  browser-agent
+
+# Customize Xvfb display settings
+docker run -p 8080:8080 \
+  -e BROWSER_XVFB_ENABLED=true \
+  -e BROWSER_XVFB_DISPLAY=:99 \
+  -e BROWSER_XVFB_SCREEN_RESOLUTION=1920x1080x24 \
+  browser-agent
+```
+
+**Security Note:** Xvfb is configured without the `-ac` flag (access control disabled) for security. The X server uses `-nolisten tcp` to prevent network access.
+
 ## License
 
 MIT License - see LICENSE file for details

agent.yaml

@@ -12,6 +12,8 @@ spec:
   config:
     browser:
       headless: true
+      engine: "chromium"
+      stealth_mode: false
       user_agent: "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36"
       viewport_width: 1920
       viewport_height: 1080
@@ -22,6 +24,9 @@ spec:
       header_connection: "keep-alive"
       header_upgrade_insecure_requests: "1"
       data_dir: "/tmp/playwright/artifacts"
+      xvfb_enabled: false
+      xvfb_display: ":99"
+      xvfb_screen_resolution: "1920x1080x24"
       args:
         - "--disable-blink-features=AutomationControlled"
         - "--disable-features=VizDisplayCompositor"

docker-entrypoint.sh

@@ -0,0 +1,60 @@
+#!/bin/bash
+set -e
+
+# Configuration from environment variables
+XVFB_ENABLED="${BROWSER_XVFB_ENABLED:-false}"
+XVFB_DISPLAY="${BROWSER_XVFB_DISPLAY:-:99}"
+XVFB_SCREEN="${BROWSER_XVFB_SCREEN_RESOLUTION:-1920x1080x24}"
+
+# Function to check if Xvfb is ready
+wait_for_xvfb() {
+    local max_attempts=10
+    local attempt=0
+
+    while [ $attempt -lt $max_attempts ]; do
+        if xdpyinfo -display "$XVFB_DISPLAY" >/dev/null 2>&1; then
+            echo "Xvfb is ready on display $XVFB_DISPLAY"
+            return 0
+        fi
+        attempt=$((attempt + 1))
+        sleep 0.5
+    done
+
+    echo "Warning: Xvfb failed to start within timeout"
+    return 1
+}
+
+# Start Xvfb if enabled
+if [ "$XVFB_ENABLED" = "true" ]; then
+    echo "Starting Xvfb on display $XVFB_DISPLAY with screen resolution $XVFB_SCREEN"
+
+    # Start Xvfb without -ac flag for security
+    # Use -nolisten tcp to prevent network access
+    Xvfb "$XVFB_DISPLAY" -screen 0 "$XVFB_SCREEN" -nolisten tcp &
+    XVFB_PID=$!
+
+    # Wait for Xvfb to be ready
+    if wait_for_xvfb; then
+        export DISPLAY="$XVFB_DISPLAY"
+        echo "Xvfb started successfully (PID: $XVFB_PID)"
+    else
+        echo "Error: Xvfb failed to start properly"
+        kill "$XVFB_PID" 2>/dev/null || true
+        exit 1
+    fi
+
+    # Trap to cleanup Xvfb on exit
+    trap "echo 'Stopping Xvfb...'; kill $XVFB_PID 2>/dev/null || true" EXIT
+else
+    echo "Xvfb disabled, using native headless mode"
+fi
+
+# Log configuration
+echo "Browser configuration:"
+echo "  Engine: ${BROWSER_ENGINE:-chromium}"
+echo "  Headless: ${BROWSER_HEADLESS:-true}"
+echo "  Stealth Mode: ${BROWSER_STEALTH_MODE:-false}"
+echo "  Xvfb Enabled: $XVFB_ENABLED"
+
+# Start the main application
+exec ./main

example/.env.example

@@ -3,5 +3,9 @@ DEEPSEEK_API_KEY=
 GOOGLE_API_KEY=
 
 # Agent
+BROWSER_ENGINE=chromium
+BROWSER_HEADLESS=false
+BROWSER_XVFB_ENABLED=true
+BROWSER_STEALTH_MODE=true
 A2A_AGENT_CLIENT_PROVIDER=deepseek
 A2A_AGENT_CLIENT_MODEL=deepseek-chat

example/Dockerfile.vnc

@@ -0,0 +1,29 @@
+FROM alpine:latest
+
+# Install x11vnc
+RUN apk add --no-cache x11vnc
+
+# Create a startup script that connects to the shared X display
+COPY <<'EOF' /usr/local/bin/start-vnc.sh
+#!/bin/sh
+set -e
+
+echo "Waiting for X display :99 to be available..."
+echo "Checking /tmp/.X11-unix directory..."
+ls -la /tmp/.X11-unix/ 2>/dev/null || echo "Directory not found or empty"
+
+while true; do
+  if [ -S /tmp/.X11-unix/X99 ]; then
+    echo "X11 socket found, attempting to start x11vnc..."
+    x11vnc -display :99 -forever -shared -passwd password -rfbport 5900 -listen 0.0.0.0 -noshm && break
+  else
+    echo "Waiting for /tmp/.X11-unix/X99 socket... ($(date))"
+    ls -la /tmp/.X11-unix/ 2>/dev/null || echo "Directory still empty"
+  fi
+  sleep 5
+done
+EOF
+
+RUN chmod +x /usr/local/bin/start-vnc.sh
+
+ENTRYPOINT ["/usr/local/bin/start-vnc.sh"]