v1.18.0

v1.18.0 - code-server (VS Code in the Browser)

Release Date: 2026-01-26

✨ New Features

🖥️ code-server - VS Code in the Browser

Full Visual Studio Code experience running in your browser with AI coding assistant support.

• AI Coding Extensions: Support for Claude Code, OpenCode, Continue, and other AI assistants
• Persistent Workspace: Files, settings, and extensions persist across sessions
• Shared Folder Integration: Direct access to AI LaunchKit's ./shared folder
• Full Terminal Access: Complete Linux terminal with sudo capabilities
• Extension Marketplace: Access to Open VSX Registry

Access: https://code.yourdomain.com

New Environment Variables:

• CODESERVER_HOSTNAME - Subdomain for code-server
• CODESERVER_PASSWORD - Login password (auto-generated)
• CODESERVER_SUDO_PASSWORD - Sudo password for terminal (auto-generated)

📋 Installation

Select "code-server" in the installation wizard under AI-Powered Development tools.

🔧 Post-Installation: Enable AI Coding Tools

Node.js for Claude Code Extension

code-server requires Node.js 18+ for the Claude Code extension. This is automatically installed via init script on container start.

Manual installation (if needed):

docker exec -it code-server bash -c "curl -fsSL https://deb.nodesource.com/setup_20.x | bash - && apt-get install -y nodejs"

Install OpenCode CLI

docker exec -it code-server npm i -g opencode-ai

Install Ollama CLI (for ollama launch)

docker exec -it code-server bash -c "apt-get update && apt-get install -y zstd && curl -fsSL https://ollama.com/install.sh | sh"

Configure OpenCode with Ollama Cloud

For cloud models (recommended for larger context windows):

1. Get your API key from https://ollama.com/settings/keys

2. Create the config file in code-server terminal:

mkdir -p /config/.config/opencode
cat > /config/.config/opencode/opencode.json << 'EOF'
{
  "$schema": "https://opencode.ai/config.json",
  "model": "ollama/glm-4.7:cloud",
  "provider": {
    "ollama": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Ollama Cloud",
      "options": {
        "baseURL": "https://ollama.com/v1",
        "apiKey": "YOUR_OLLAMA_API_KEY"
      },
      "models": {
        "glm-4.7:cloud": {
          "name": "GLM 4.7 Cloud"
        }
      }
    }
  }
}
EOF

3. Start OpenCode:

opencode

Configure OpenCode with Local Ollama

For local models via AI LaunchKit's Ollama container:

mkdir -p /config/.config/opencode
cat > /config/.config/opencode/opencode.json << 'EOF'
{
  "$schema": "https://opencode.ai/config.json",
  "model": "ollama/qwen2.5:7b-instruct-q4_K_M",
  "provider": {
    "ollama": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Ollama Local",
      "options": {
        "baseURL": "http://ollama:11434/v1"
      },
      "models": {
        "qwen2.5:7b-instruct-q4_K_M": {
          "name": "Qwen 2.5 7B"
        }
      }
    }
  }
}
EOF

n8n-MCP for Workflow Generation

If you have n8n-MCP enabled in AI LaunchKit, you can install the MCP client in code-server:

docker exec -it code-server npm i -g n8n-mcp

Configuration for OpenCode/Claude Code:

# In code-server terminal:
cat >> ~/.bashrc << 'EOF'
export N8N_MCP_URL="http://n8nmcp:3000"
export N8N_MCP_TOKEN="your-token-from-env"
export N8N_API_URL="http://n8n:5678"
export N8N_API_KEY="your-n8n-api-key"
EOF
source ~/.bashrc

This enables AI assistants to generate n8n workflows directly!

🔗 Integration with AI LaunchKit Services

Service	Access from code-server
n8n	http://n8n:5678
n8n-MCP	http://n8nmcp:3000
Ollama (local)	http://ollama:11434
Ollama Cloud	https://ollama.com/v1
Gitea	http://gitea:3000
Shared Folder	/config/workspace/shared

📚 Documentation

Full documentation added to README.md including:

• Initial setup guide
• AI extension configuration (Claude Code, OpenCode)
• Ollama Cloud and local model configuration
• n8n-MCP integration for workflow generation
• Troubleshooting guide

---

Full Changelog: v1.17.2...v1.18.0