CodeCleaner AI is designed to make you a more effective and confident developer. Here’s why it’s a great addition to your toolkit:

• Boost Your Code Quality: Get AI-powered suggestions to write cleaner, more reliable code and identify potential problems before they escalate.
• Solve Bugs Faster: Quickly understand error messages and complex code, with clear guidance on how to fix issues efficiently.
• Learn and Grow: CodeCleaner AI doesn’t just give answers; it explains the ‘why’ behind them, helping you master concepts and improve your coding skills with every session.
• Maintain Control and Privacy: You always decide what to implement based on CodeCleaner AI suggestions.

CodeCleaner AI helps you ship better code and become a stronger developer, no matter your current experience level.

Absolutely! CodeCleaner AI is a fantastic tool if you’re learning to code. Think of it as your personal AI tutor, designed to help you:

• Understand Tricky Concepts: When you’re puzzling over why your code isn’t working or what an error message really means, CodeCleaner AI can provide clear explanations.
• Fix Bugs with Guidance: It helps you understand the steps to fix issues, not just a quick fix, so you learn along the way.
• Bridge Theory and Practice: Get real-time feedback and explanations on the code you’re actually writing, helping you connect what you’ve learned to real-world application.
• Build Confidence: By getting support when you’re stuck, you can learn faster and feel more confident in your coding journey.

It’s perfect for those moments when you wish you had an expert by your side to ask “Why isn’t this working?” or “What does this error mean?”. You can easily get started and explore these benefits with our free Starter Plan.

Definitely. For experienced developers, CodeCleaner AI acts as a powerful AI pair-programmer. It’s designed to augment your expertise and help you:

• Accelerate Debugging: Get to the root of complex bugs and unexpected behaviors more quickly, saving you valuable time.
• Master Legacy Code: Rapidly understand, navigate, and identify areas for improvement or expansion in existing or unfamiliar codebases.
• Enhance Code Quality: Use it to spot subtle issues, potential vulnerabilities, or opportunities for optimization that can be easy to overlook in complex projects.
• Validate Solutions: Get an intelligent “second opinion” on your approach when implementing new features or making significant changes.

Think of it as a way to boost your productivity, tackle challenging code with more insight, and ensure you’re shipping robust, high-quality software. Our Developer Plan is tailored to provide the power and flexibility you need.

Absolutely. For expert developers, CodeCleaner AI acts as a sophisticated assistant, helping you enhance critical oversight and strategic tasks. It can be particularly valuable for:

• Streamlining Code Reviews: Accelerate your review process by having AI flag potential bugs, security vulnerabilities, and deviations from best practices. This allows you to focus your expertise on architectural soundness and strategic implementation.
• Ensuring Production Readiness: Get an AI-powered analysis to help confirm if code meets the highest standards for quality, robustness, and security before deployment.
• Deep System Understanding: Quickly gain insights into complex or unfamiliar sections of code, helping you make informed decisions for refactoring or expansion.
• Upholding Best Practices: Use its analysis as a tool to identify patterns, enforce coding standards, and even support mentoring efforts within your team by highlighting areas for improvement.

Our Power User Plan, which includes features like GitHub integration, is specifically designed to provide the advanced capabilities and efficiency that expert-level work demands.

Creating your CodeCleaner AI account is simple:

1. Go to app.codecleaner.ai/signup.
2. You can sign up by entering your email address and creating a password, or you can choose the “Sign up with Google” option for a quicker setup.
3. If you sign up with email, you’ll need to verify your email address.
4. Once completed, your account will be created, and you’ll automatically start on our free Starter Plan, ready to explore CodeCleaner AI!

Once your account is created:

1. On your desktop or laptop computer, open a modern web browser. CodeCleaner AI is designed for detailed code interaction, so for the best experience, we recommend using a device with a larger screen.
2. Navigate to app.codecleaner.ai.
3. Log in with your email and password, or use the Google account you signed up with.

Once you’re logged in, you’ll land right on your CodeCleaner AI dashboard. Time to dive in and make your code shine!

CodeCleaner AI is designed to provide the best experience on desktop and laptop computers.

While you might be able to access our web application on a mobile device’s browser, it’s not currently optimized for smaller screens. You’ll likely find detailed code interaction challenging.

For full functionality and a comfortable user experience, we strongly recommend using CodeCleaner AI on a larger screen.

No, you don’t need to install any software to use CodeCleaner AI! It’s a fully web-based application.

All you need is an internet connection and a modern web browser on your desktop or laptop, and you can access it directly by navigating to app.codecleaner.ai and logging in.

You don’t need a powerful machine to use CodeCleaner AI, as it’s a web application. The main things you’ll want to have are:

• A Modern Web Browser: We suggest using an up-to-date version of common browsers like Google Chrome, Mozilla Firefox, Safari, or Microsoft Edge.
• A Stable Internet Connection: This is needed to use the app and for the AI to provide its analysis and suggestions.
• A Desktop or Laptop Computer: CodeCleaner AI is designed for detailed work with code, so it works best and is most comfortable to use on a larger screen.

That’s essentially it. If your computer can run a modern web browser smoothly, you should be good to go.

CodeCleaner AI is designed to be highly versatile and support a vast range of programming languages and frameworks. This is possible because we leverage advanced Large Language Models (from providers like OpenAI, Gemini, and Anthropic) that have been trained on an enormous amount of code.

This means you can confidently use CodeCleaner AI with:

• Popular languages: Python, JavaScript (and its frameworks like React, Angular, Vue), Java, C#, C++, Go, Ruby, PHP, Swift, Kotlin, TypeScript, SQL, and many more.
• Many other languages: The underlying AI models understand the structure and patterns of most programming languages, even if they are less common or more specialized.

While it’s practically impossible to officially test and list every single language, the core technology is built for broad linguistic understanding. So, if you’re working with a language, there’s a very strong likelihood CodeCleaner AI can assist you. We encourage you to try it with your specific stack!

Using CodeCleaner AI is an interactive process structured around helping you with your coding tasks. Here’s a general idea:

1. Define Your Context with a “Project”: You’ll typically start by setting up a “Project.” This tells CodeCleaner AI key details like the programming language you’re using and what your code is about. Providing this context helps the AI give you more tailored and accurate advice.
2. Focus Your Work in a “Session”: When you have a specific piece of code to analyze, a bug to fix, or a concept to understand, you’ll do this within a “Session” linked to your project.
3. Interact Using “Actions”: Within a session, you use different “Actions” to ask for help. This could be a “Code Action” if you have a specific snippet (e.g., asking the AI to ‘Explain’ or ‘Debug’ it), or a “Project Action” for broader guidance.
4. Receive AI Feedback & Iterate: CodeCleaner AI processes your request and provides explanations, suggestions, or analysis. You can then use “Followup Actions” (like ‘Clarify’ or ‘Challenge’) to refine the conversation and deepen your understanding.
5. You Implement Changes: CodeCleaner AI guides and suggests, but you always remain in control of making any actual changes to your codebase.

For more details on these key components, please see our FAQs on “What is a Project?”, “What is a Session?”, and “What are Actions?”.

In CodeCleaner AI, a “Project” is how you organize your work and provide essential context to our AI. Think of it as creating a dedicated workspace for a specific coding task or application you’re developing.

When you set up a Project, you’ll typically define key details such as:

• Project Name: A clear and descriptive name (e.g., “User Authentication Module,” “Inventory Management Script”).
• Description: A brief summary of what the project is about or its main goals.
• Programming Language: The primary language being used (e.g., Python, JavaScript, Java).
• Other relevant information that helps define your work.

Supplying these Project details is useful. It helps CodeCleaner AI get a better understanding of what you’re working on. That way CodeCleaner AI can give you suggestions that are based on your context instead of trying to guess what you are working on.

Think of a “CodeCleaner AI Session” as your focused, one-on-one conversation with our AI about a specific coding task, question, or problem you’re working on. It’s the dedicated space where you interact with the AI to get help.

Here’s what’s typical for a Session:

• It’s Tied to a Project: Each Session happens within one of your “Projects.” This means the AI uses your Project’s details (like the programming language and project description) to give you relevant advice.
• It’s an Interaction: A Session consists of a series of exchanges. You use “Actions” to ask questions or provide code, and CodeCleaner AI responds with explanations or suggestions. You can then use “Followup Actions” to continue the dialogue.
• It’s for Specific Goals: You might start a Session to understand a particular piece of code, debug an error, get suggestions for a new feature, or have comments added to a function.

Essentially, each Session is like an active chat dedicated to helping you with one specific item within your broader Project.

In CodeCleaner AI, “Actions” are basically how you tell the AI what you need help with or what you want it to do. Think of them as the different types of requests you can make. There are three main categories:

• Code Actions: You’ll use these when your question or task is about a specific piece of code within your current “Session.” For example, you might select your code and use a Code Action to ask the AI to ‘Explain’ it, help ‘Debug’ an error, ‘Comment’ it, or ‘Suggest a Fix.’

• Project Actions: These are for broader questions or guidance related to your overall “Project” rather than a single code snippet. You might use a Project Action to get ‘Guidance’ on your project’s structure or discuss how to ‘Do More’ with it, like planning new features.

• Followup Actions: After CodeCleaner AI provides an initial response to one of your Code or Project Actions, you use Followup Actions to continue the conversation. This allows you to ‘Clarify’ something, ‘Challenge’ a suggestion, ‘Check’ if your own solution is correct, or simply have a more unstructured chat to explore the AI’s feedback further.

Each type of Action helps you get the specific kind of assistance you need from CodeCleaner AI.

“Bring Your Own Key,’ or BYOK, is a feature that allows you to use CodeCleaner AI with your own personal or company API keys from the Large Language Model (LLM) providers we support.

Instead of CodeCleaner AI using a central, shared key to access these AI models (like those from OpenAI, Google for its Gemini models, or Anthropic), you provide your own API key directly within our application. This essentially connects CodeCleaner AI to your existing accounts with these AI providers, using your own access credentials.

We use a “Bring Your Own Key” (BYOK) model because it offers you, the user, several significant advantages:

• Full Control Over Your AI Model Costs: When you use your own API keys from providers like OpenAI, Google (for Gemini models), or Anthropic, you pay them directly for the AI model usage. This means you can benefit from any specific pricing, free tiers, or credits you have with these providers. CodeCleaner AI doesn’t add any markup to these costs.
• Flexibility and Access to More AI Models: With your own key, you often have the flexibility to choose from a wider range of AI models offered by the provider, including their latest releases, as they become available on your account.
• Manage Your Own Usage Limits: Your API usage is governed by the rate limits and quotas of your own account with the LLM provider. This gives you direct control and visibility over your consumption.
• Clear Data Handling with LLM Providers: Using your own API key means that when your data is sent to the LLM provider for processing, it’s done under your account and subject to your agreement with them. This provides a clear pathway for how that data is handled by the LLM. (And as a reminder, CodeCleaner AI itself does not use your code to train any AI models).
• Transparent Subscription Value: The BYOK approach allows us to keep the CodeCleaner AI subscription focused on the features and value our application provides, rather than bundling or reselling fluctuating AI model costs.

Ultimately, our BYOK model is designed to give you greater transparency, choice, and control when leveraging powerful AI models to help with your code.

Setting up your own API keys (BYOK) to use with CodeCleaner AI is a straightforward process that allows you to connect to your preferred Large Language Model providers. The general steps involve getting an API key from the provider and then adding it to your CodeCleaner AI account settings.

For detailed, step-by-step instructions, complete with screenshots and guidance for each supported LLM provider, please see our comprehensive guide on our documentation site:

https://docs.codecleaner.ai/docs/key-concepts/bring-your-own-key

That page will walk you through everything you need to know to get your keys configured and ready to use with CodeCleaner AI.

The ‘Assess’ Code Capability in CodeCleaner AI acts like an initial health check for your code. When you use this feature on a specific code snippet, our AI analyzes it to give you an overview of its quality and structure.

Specifically, ‘Assess’ looks for things like:

• Potential logical errors or common bugs.
• Whether the code follows general best practices and coding standards.
• Areas that might be confusing, overly complex, or could be made more efficient.
• Possible security considerations or vulnerabilities.

After its analysis, CodeCleaner AI provides you with a summary of observations, pointing out both strengths and areas that might need improvement or further attention. It’s a great way to get a quick, high-level understanding of your code’s condition before you start making changes or diving into specific debugging.

The ‘Explain’ Code Capability in CodeCleaner AI is designed to help you understand any piece of code better. Whether you’re looking at a complex algorithm, an unfamiliar function, or code written in a language you’re still learning, this feature can provide clarity.

When you use ‘Explain’ on a selected code snippet, CodeCleaner AI will:

• Describe the code’s overall purpose and what it’s trying to achieve.
• Break down its logic step-by-step, if applicable.
• Clarify how different parts of the code work together.
• Explain specific programming constructs or syntax if they are central to understanding that snippet.

This makes the ‘Explain’ feature useful for learning new programming concepts, for figuring out how unfamiliar or legacy code works, or when you simply need to understand a code snippet quickly before you make changes or build on it.

The ‘Comment’ Code Capability in CodeCleaner AI helps you quickly add explanatory comments directly into your code, making it easier to understand and maintain.

When you use the ‘Comment’ feature on a code snippet you provide, CodeCleaner AI will:

1. Analyze that code.
2. Then, it provides you with a new version of that same code, but enriched with inline comments.
3. These comments are added using the correct syntax for your code’s programming language and aim to explain what specific blocks or sections of your code are doing.

You’ll receive this commented version of your code as a suggestion from CodeCleaner AI. You can then review these added comments and, if you find them helpful, copy this commented code to use in your project. CodeCleaner AI doesn’t automatically change your original files; it gives you this enhanced version for you to use as you see fit.

The ‘Debug’ Code Capability in CodeCleaner AI is designed to help you understand and resolve specific errors you encounter while coding. If you’re faced with a confusing error message, this feature not only explains the error but can also suggest a code fix.

To use ‘Debug’, you’ll typically provide two pieces of information:

1. The Error Message: The exact error message you’re seeing.
2. The Code (Optional): The specific block of code that you believe is causing the error. Providing the code is optional but is highly recommended, as it allows CodeCleaner AI to give you much more targeted advice and a more relevant suggested code fix.

Once you provide this information, CodeCleaner AI will analyze it and aim to provide you with:

• An explanation of what the error message likely means in simpler terms.
• Insights into potential causes of the error, based on the message and your code (if provided).
• A suggested version of your code snippet with modifications intended to correct the bug.

This capability helps you troubleshoot explicit errors more efficiently by not only explaining the problem but also by offering a direct code suggestion to resolve it. As with all CodeCleaner AI features, this corrected code is a suggestion. You review it, understand the changes, and then decide whether to implement it in your actual codebase.

The ‘Fix’ Code Capability in CodeCleaner AI is designed to help you when your code runs without throwing an error, but it doesn’t produce the output or behavior you intended. If your code isn’t working the way it’s supposed to, ‘Fix’ can help identify the logical issue and suggest a corrected version of your code.

To use this feature most effectively, you’ll typically need to provide:

1. The Code: The specific snippet of code that is producing the incorrect behavior.
2. Expected Outcome: A clear description of what you wanted the code to do or what result you expected.
3. Actual Outcome: A description of what the code is actually doing or the incorrect result it’s currently producing.

Based on this information—your code, what you wanted, and what you got—CodeCleaner AI will analyze the logic. It then aims to provide you with:

• An explanation of why your code might not be achieving the expected outcome.
• A suggested version of your code with modifications intended to produce the correct, expected behavior.

This capability helps you correct logical flaws and get your code to function as intended. The revised code provided by CodeCleaner AI is a suggestion; you’ll review it to understand the changes and then decide whether to implement them in your project.

The ‘Guidance’ Project Capability in CodeCleaner AI provides you with high-level advice, direction, and strategic insights for your overall software project. Unlike “Code Actions” that focus on specific snippets of code, ‘Guidance’ helps you think about your project more broadly.

When you use this capability, CodeCleaner AI draws upon the context you’ve defined in your “Project” (such as its description, goals, and programming language) to offer suggestions like:

• Potential architectural approaches or design patterns to consider.
• Ideas for how to plan or structure the implementation of new features.
• General best practices relevant to your project’s type or technology stack.
• Things to consider or potential roadblocks you might encounter based on your project description.

Think of the ‘Guidance’ capability as a way to get a “second opinion” or brainstorm with an AI assistant about your project’s development path, helping you make more informed strategic decisions.

The ‘Clarify’ Followup Action is what you use when you need CodeCleaner AI to better explain something it said in a previous response. If part of the AI’s explanation, suggestion, or analysis was unclear, or if you just want more detail on a specific point, ‘Clarify’ helps you ask for that.

When you use this action, you’ll typically specify what you found confusing or what you’d like more information about from the AI’s last message. In response, CodeCleaner AI will try to:

• Rephrase the particular point in a different way.
• Provide more details or a deeper explanation.
• Break down a complex idea into simpler parts.

The goal of the ‘Clarify’ action is to help you get a complete and clear understanding of the AI’s feedback, so you can use its insights effectively.

The ‘Challenge’ Followup Action is a powerful tool you can use when you want to question, disagree with, or explore alternatives to a specific point made by CodeCleaner AI in its previous response. If you think the AI’s suggestion might not be the best approach, if its explanation seems incomplete, or if you simply have a different idea you’d like it to consider, ‘Challenge’ is how you can express that.

When you use this action, you’ll typically specify the part of the AI’s last message you are questioning and provide your counterpoint, reason for disagreement, or alternative suggestion. CodeCleaner AI will then:

• Re-examine its previous statement in light of your challenge.
• Try to provide further justification for its original point, if it still believes it’s valid.
• Offer revised suggestions or alternative viewpoints if your challenge brings up a good point or a different angle.
• Acknowledge any limitations or potential inaccuracies in its earlier response if your challenge highlights them.

Using the ‘Challenge’ action fosters a more dynamic and critical interaction, helping to refine the AI’s output and potentially leading to better, more nuanced solutions or understanding. It encourages you to think critically alongside the AI.

The ‘Validate’ Followup Action is a tool you use during an ongoing CodeCleaner AI session, right after our AI has provided some initial assistance (like helping you ‘Debug’ an error or ‘Explain’ a concept). Once you’ve taken that feedback and drafted your own code solution based on that specific discussion, ‘Validate’ lets you ask the AI to review your attempt. It’s your way to check if your approach is on track, if you’ve correctly applied earlier advice, or if your new code itself has an issue.

Here’s how you provide information for the ‘Validate’ action:

1. Your Code Solution (Required): You must provide the code snippet you’ve written that you want the AI to examine. This should be based on the current session’s topic.
2. Error Message from Your Solution (Optional): If, after writing your code solution (from step 1), you tested it and it produced an error, you can include that specific error message here.
3. Problematic Code Section (Optional, and only if you also provided an Error Message in step 2): If you’ve included an ‘Error Message from Your Solution’, you can then also highlight the specific part of your code solution (from step 1) that you believe is causing this error. This helps the AI focus its debugging analysis. You cannot provide this if you haven’t provided an error message.

CodeCleaner AI then analyzes your inputs in the context of your ongoing session:

• If you provide just ‘Your Code Solution’: The AI will assess if it seems to address the problem or goal discussed in the session, check if it aligns with previous advice, identify potential issues or improvements, and comment on your logic.
• If you provide ‘Your Code Solution’ AND an ‘Error Message from Your Solution’: In addition to the above, the AI will help explain the error and suggest how your code solution might be modified to fix it.
• If you provide all three inputs (including ‘Problematic Code Section’): The AI will do all of the above, but its suggestions for fixing the error in your code solution will pay special attention to the ‘Problematic Code Section’ you highlighted.

So, using the ‘Validate’ action is a good way to:

• Try out your own coding ideas.
• Get specific feedback if your solution isn’t quite working.
• Check if you’re generally heading in the right direction.
• Learn as you go by discussing your code with the AI during your session.

The ‘Chat’ Followup Action in CodeCleaner AI allows you to have a more open-ended, free-form conversation with the AI based on the current session’s topic. It’s different from more specific followups like ‘Clarify’ (which asks for a clearer explanation of a previous point), ‘Challenge’ (which questions a previous point), or ‘Validate’ (which checks your code solution).

You might use the ‘Chat’ action when you want to:

• Discuss an idea or a related concept that came up during your session.
• Ask broader “what if” questions about the code or problem you’re working on.
• Explore the AI’s reasoning or suggestions in a more conversational way.
• Brainstorm alternative approaches or delve into nuances not covered by the more structured actions.

Think of the ‘Chat’ followup as opening a direct line to talk more generally with CodeCleaner AI about the subject at hand, allowing for flexible exploration and a deeper, more interactive learning experience beyond the specific structured prompts.

No, CodeCleaner AI will never automatically modify your original code files in your editor or project. Our philosophy is to act as your AI-powered assistant, providing suggestions and insights while keeping you firmly in control.

When you use capabilities like ‘Comment,’ ‘Debug,’ or ‘Fix’ that involve changes to code:

• CodeCleaner AI will present you with a suggested version of your code (e.g., with added comments, or with proposed fixes).
• This is always presented as a suggestion within the CodeCleaner AI interface.

It is then entirely up to you, the developer, to review these suggestions, understand them, and then decide if you want to copy and paste them, or manually apply the proposed changes, into your actual project files. We believe this approach empowers you to learn from the insights and maintain full authority over your codebase.

Yes, you can easily use and save the suggestions, explanations, and code that CodeCleaner AI provides. We’ve built in convenient ways for you to do this directly from our interface.

When CodeCleaner AI gives you a response:

• You’ll typically find an option to copy the full response from the AI. This is handy if you want to save the entire piece of advice or explanation for your notes or documentation.
• For any code snippets within the AI’s response (like a suggested fix, a commented version of your code, or an example), there are usually specific buttons or options that allow you to copy just that code block with a single click.

Once copied, you can then paste this information directly into your code editor, project files, personal notes, or anywhere else you need it. This makes it straightforward to integrate the AI’s assistance into your development workflow.

Yes, CodeCleaner AI does keep a history of your interactions within each “Session.” Here’s how it works:

• Organized by Project: Your work in CodeCleaner AI is organized into “Projects.” All the “Sessions” you conduct are stored within their respective Project.
• Accessing Session History: For each session you’ve had, you’ll typically find a way to view its history (for example, by clicking on the session name or a specific link for it within your Project).
• Chat-like Display: When you open a session’s history, it will display the full sequence of interactions – your questions or “Actions” and CodeCleaner AI’s responses – presented as a series of chat messages.

This history allows you to easily go back and review previous discussions, recall solutions or explanations the AI provided, and track the evolution of your work on a particular issue.

Yes, CodeCleaner AI offers you the flexibility to switch between different Large Language Model (LLM) providers (like OpenAI, Google for its Gemini models, or Anthropic) and even specific models offered by those providers, often at any point during an active “Session.”

Provided you have set up your API keys for these providers using our “Bring Your Own Key” (BYOK) system, you can typically change your selection for the AI model that CodeCleaner AI will use for its next interaction with you.

This capability is useful because it allows you to:

• Tailor the AI to Your Task: You might choose a highly advanced model for complex problem-solving or a faster, more cost-effective model for simpler questions or explanations.
• Experiment and Compare: See which model or provider gives you the most helpful responses for different types of coding challenges or questions.
• Optimize for Cost and Performance: Balance your needs for AI capability with the cost associated with using different models via your own API keys.
• Leverage Specific Strengths: Different models or providers might have unique strengths, and you can switch to take advantage of them as needed.

You’ll generally find the option to change the active LLM and model within the session interface, allowing you to adapt the AI assistance dynamically.

Yes, there are limitations on the total amount of code you can process with CodeCleaner AI, and these generally depend on your subscription plan.

This is often referred to as an “Input Limit” and is typically measured by the number of lines of code you submit for analysis, explanation, debugging, fixing, or other actions over a certain period (usually monthly). Each of our plans—Starter, Developer, and Power User—comes with a different allowance for this.

To find the specific details on the lines of code limits for each subscription plan, please see our Plans page. This will provide a clear breakdown of the processing capacity included with each plan, helping you choose the one that best fits your needs.

CodeCleaner AI is designed to give you flexibility and control over the AI power you use. Through our “Bring Your Own Key” (BYOK) feature, you can connect CodeCleaner AI to your own accounts with several leading Large Language Model (LLM) providers.

As of now, the primary LLM providers we support are:

• OpenAI (provider of models like the GPT series)
• Google (provider of models like Gemini)
• Anthropic (provider of models like Claude)

By setting up your API keys from these providers in CodeCleaner AI, you can choose which AI engine powers your sessions. This allows you to leverage the specific strengths, performance characteristics, and pricing models of each provider and their various models.

Yes, generally you can, and CodeCleaner AI is designed to adapt to the fast-moving world of Large Language Models (LLMs). Here’s our approach:

• Flexibility via “Bring Your Own Key” (BYOK): Our BYOK system is a key advantage. Often, when LLM providers like OpenAI, Google (for Gemini models), or Anthropic release new models or updates accessible via your existing API key, you might be able to start using them with CodeCleaner AI fairly quickly, as you’re using your direct access to the provider.
• Regular Platform Updates: We continuously monitor new developments from the LLM providers. Our team works to regularly update CodeCleaner AI to:
  • Ensure ongoing compatibility with model changes.
  • Explicitly list new, well-tested, and beneficial models within our interface’s model selection choices, making them easy for you to find and use.
• Focus on Value and Stability: While we aim for broad support, we also prioritize integrating models that demonstrate stable performance, reliability, and clear value for coding-related tasks.
• Where to Find the Latest Options: The most up-to-date list of models explicitly selectable within CodeCleaner AI can usually be found in the model selection dropdown when you are setting up or changing models within a session (assuming your API key for that provider is active). We also announce significant new model integrations through our product updates or community channels.

Our goal is to provide you with access to a powerful and relevant range of AI models, balancing cutting-edge capabilities with reliability, all through the flexible BYOK system.

Whether we add support for additional LLM providers in the future really depends on user feedback and demand. We believe in building CodeCleaner AI in a way that best serves our community’s needs.

If there’s a specific LLM provider or a new model you’re particularly interested in using with CodeCleaner AI, or if you have ideas about how different AI capabilities could improve the platform, we’d love to hear from you! The best place to share these suggestions, discuss them with our team, and connect with other CodeCleaner AI users is on our Discord server.

In the context of CodeCleaner AI and our “Bring Your Own Key” (BYOK) model, an API key is the unique code you obtain from an LLM provider (such as OpenAI, Google for its Gemini models, or Anthropic). This key allows CodeCleaner AI, acting with your permission, to send requests to that provider’s AI models using your account and receive their responses.

When CodeCleaner AI uses your API key, its main purposes are to:

• Authenticate your requests, confirming to the LLM provider that the calls are authorized by you.
• Enable the LLM provider to track your API usage for billing directly to your account with them.
• Ensure your usage adheres to any rate limits or access permissions associated with your specific key and account.

API keys are sensitive credentials as they grant access to your LLM provider accounts and are often linked to billing. We take the security of any API keys you add to CodeCleaner AI very seriously and use robust security measures to protect them within our system. (You can learn more about our specific security practices in our ‘Data Privacy & Security’ FAQ section or by reviewing our Security Policy). On your part, it’s also essential to treat your API keys with the same level of care as a password, keeping them confidential.

To use CodeCleaner AI with our “Bring Your Own Key” (BYOK) model, you’ll need to get an API key directly from the Large Language Model (LLM) provider whose AI models you wish to use. This usually means you’ll need to create an account on their platform first if you don’t already have one.

Here’s general guidance for the LLM providers currently supported by CodeCleaner AI:

• OpenAI (for models like the GPT series): Typically, you generate and manage API keys after logging into your account on the OpenAI Platform website, often in an “API keys” section of your settings or developer dashboard.

• Google (for Gemini models): For Gemini API keys, you’ll generally need to use Google AI Studio or the Google Cloud Console. Look for sections related to “Generative AI,” “Vertex AI,” or API key management within your Google account.

• Anthropic (for models like Claude): Log in to your account on the Anthropic Console website, where you should find options to create and manage your API keys.

Important Note on Instructions: The exact steps and user interfaces on these provider websites can change. You can find helpful guides and links to these provider resources on our documentation site at https://docs.codecleaner.ai.

While our guide can help you get started, we also always recommend referring to the official, most current documentation provided directly by OpenAI, Google, or Anthropic, as they are the ultimate source of truth for their services. Also, remember that using these API keys subjects you to the terms and billing policies of that specific LLM provider.

Once you have your API key, you can add it to CodeCleaner AI (see our FAQ: “How do I set up and use my own API keys with Code Cleaner AI?”).

When you add your API key from an LLM provider (like OpenAI, Google for its Gemini models, or Anthropic) to CodeCleaner AI using our “Bring Your Own Key” (BYOK) system, we use that key specifically to enable our platform to interact with your chosen AI model on your behalf.

Here’s how your API key is typically used in the process:

1. Initiating an AI-Powered Action: When you select an action in CodeCleaner AI that needs AI assistance (for instance, asking to ‘Explain Code,’ ‘Debug,’ or ‘Fix’), our application prepares the necessary information, including your code and context from your “Project” and “Session.”
2. Authorizing Communication with the LLM: CodeCleaner AI then uses your provided API key to securely send this prepared request to your selected LLM provider. Your API key acts as your credential, authenticating the request and authorizing the provider to process it under your account.
3. Linking Usage to Your Account: The LLM provider tracks any AI model usage from these requests as activity on your account. They do this to manage your usage based on your quotas and for any billing, according to your agreement with them.

Your API key is essential for authorizing these AI-powered interactions with the LLM provider. CodeCleaner AI uses your key strictly for this purpose—to facilitate the features you activate within our application.

Protecting the security and privacy of your code is a fundamental commitment at CodeCleaner AI. We use a combination of technical safeguards and strict policies to ensure your code is handled responsibly:

1. Secure Transmission:

   • All data, including your code, that’s exchanged between your browser, the CodeCleaner AI platform, and any connected Large Language Model (LLM) providers (when you use our “Bring Your Own Key” - BYOK feature) is encrypted in transit using industry-standard HTTPS/TLS.

2. Code Storage and Protection by CodeCleaner AI:

   • To allow you to access your session history or restart previous sessions, CodeCleaner AI stores the code snippets and interactions from your sessions for up to 365 days.
   • All of this session data, including your code, is encrypted at rest on our servers using strong encryption standards to protect its confidentiality.

3. Interaction with LLM Providers via BYOK:

   • When you use your own API key for an LLM provider (like OpenAI, Google for its Gemini models, or Anthropic), your code snippets and prompts are sent to that provider for processing. This interaction is governed by your API key and your direct agreement (including their data usage and privacy policies) with that LLM provider. CodeCleaner AI ensures the secure transmission of this data to the provider.

4. Strictly Controlled Internal Access to Your Code:

   • Access to your code by Wermi Pte Ltd (the company behind CodeCleaner AI) employees is highly restricted.
   • A limited number of authorized team members may only access your code if you explicitly request assistance for a support issue, and only after we have received your documented consent to do so for that specific purpose.
   • We operate on the principle of least privilege. This means our team members are only granted the minimum level of data access that is absolutely necessary to perform their specific job duties, particularly when it comes to potentially accessing user data for support. This is enforced through technical and administrative controls to safeguard your code.

5. Your Code is Not Used for AI Training by Us:

   • As we’ll detail in a separate FAQ, CodeCleaner AI does not use your code to train our own AI models or any general third-party AI models. If you use the BYOK feature, the handling of your data by the LLM provider is subject to their policies.

We believe in transparency regarding how your data is handled. For complete details, please always refer to our official Privacy Policy and Terms of Service.

No, CodeCleaner AI itself does not use your code to train any AI models. We are committed to the privacy of your code.

All of CodeCleaner AI’s features that provide AI-powered assistance operate using our “Bring Your Own Key” (BYOK) model. This means:

• When you use an AI feature in CodeCleaner AI, you are connecting to an external Large Language Model (LLM) provider (such as OpenAI, Google for its Gemini models, or Anthropic) using your own personal or company API key.
• Your code snippets and any related prompts are sent by CodeCleaner AI directly to the LLM provider you have selected and configured with your key.
• Consequently, how your data is handled by that LLM provider—including whether it might be used to train their AI models—is determined entirely by your direct agreement with that LLM provider and their specific data usage policies.

We strongly encourage you to review the terms of service and data privacy policies of any LLM provider you choose to connect to CodeCleaner AI via your API key. Many providers offer specific terms regarding data submitted through their APIs, and some may offer controls or opt-out mechanisms concerning the use of data for training their models.

To summarize: CodeCleaner AI does not use your code for AI model training. When your code is processed by an LLM provider through your own API key, that provider’s policies on data usage and training will apply.

You do. You retain ownership of the code you write and the modifications you make when using CodeCleaner AI as an assistive tool.

CodeCleaner AI is designed to be an AI-powered assistant or pair-programmer. It provides suggestions, explanations, or modified versions of code based on your inputs and interactions. However, you are the one who reviews these outputs, makes the decision to use them, and ultimately integrates them into your codebase.

Think of CodeCleaner AI like any other development tool, such as a code editor, a compiler, or a linter. While the tool helps you in the creation or modification process, the intellectual property and ownership of the final work product (your code) remain with you. This applies to both the original code you bring to the platform and any new code you develop with the assistance of CodeCleaner AI’s suggestions.

For the definitive statement on this, please always refer to our official Terms of Service. If you are using the “Bring Your Own Key” (BYOK) feature with third-party LLM providers, their terms of service regarding content ownership for the output generated via their APIs would also apply to your interactions with those specific models.