Frequently Asked Questions

Find answers to common questions about installing, using, and getting the most out of Cline. View full documentation →

Getting Started

Cline is an AI-powered coding assistant that runs in VS Code. It goes beyond simple autocompletion by reading and writing across multiple files, executing commands, and adapting to your workflow—like having a skilled developer pair-programming with you right inside your editor.

Cline understands your codebase context and can help with everything from small code edits to complex refactoring tasks.

  1. Open VS Code
  2. Click the Extensions icon in the Activity Bar
  3. Search for "Cline" and click Install
  4. Once installed:
    • Click the Cline icon OR
    • Use CMD/CTRL + Shift + P and type "Cline: Open In New Tab"
  5. Add an API key to start using Cline

For detailed instructions, check out our documentation.

When you create your Cline account, you get instant access to frontier AI models:

  • Anthropic Claude 3.7 Sonnet (recommended for coding)
  • DeepSeek Chat (cost-effective alternative)
  • Google Gemini 2.0 Flash
  1. Click the 'Sign In' button in the Cline extension
  2. Create your account at app.cline.bot
  3. Paste your API key
  4. Select your preferred model
  • Deep Context Understanding: Cline reads your entire codebase to provide more accurate, contextual assistance
  • Multi-File Edits: It can read, refactor, and update multiple files at once for more holistic changes
  • Agentic Capabilities: Cline can run tests, open a browser, or perform deeper tasks than typical AI autocomplete tools
  • Permission-Based: You control and approve any file changes or command executions
  • Plan & Act Modes: Cline can plan complex solutions before implementing them, giving you more control over the development process

Yes, Cline itself is completely free to use. You only pay for token access from a wide selection of frontier AI models.

We offer Cline credits that make it easier to experiment with different models without the hassle of sourcing and managing multiple API keys from different providers.

Cline supports a wide range of programming languages including Python, Java, C#, JavaScript/TypeScript, Go, Rust, and many more.

Since it leverages powerful AI models' understanding, new or lesser-known languages may also work, depending on the model's training. If you're working with a specialized language, try it out—you might be pleasantly surprised by Cline's capabilities.

Using Cline

  1. Open Cline in VS Code:
    • Via the Cline icon OR
    • Use CMD/CTRL + Shift + P -> "Cline: Open In New Tab"
  2. Type your request in natural language—Cline understands plain English
  3. For complex tasks, use this pattern:
    • Start with your goal: "I want to build..."
    • Provide context: "Here's what I have so far..."
    • Ask for specific actions: "Can you help me..."

Tell Cline which file(s) you want to edit:

  • Use path: "Please edit /{RELATIVE-PATH-TO}/index.js"
  • OR use @: "@index.js"

Cline will ask for confirmation before making changes (unless auto-approve settings are enabled)

Yes! Cline will confirm before running sensitive commands (like downloading large packages)

Yes. Cline is designed with privacy in mind. Your code never leaves your machine unless you connect to an external AI API.

Even then, you control exactly what is sent to the AI model. You can use tools like .clinerules to set project-specific guidelines, and Cline's permission-based approach ensures you approve any actions that might affect your code.

Cline's Plan & Act mode provides a two-stage approach to complex tasks:

  • Plan Mode: Discuss your requirements with Cline, brainstorm solutions, and develop a detailed implementation plan
  • Act Mode: Once you approve the plan, Cline executes it step by step, making the necessary code changes and running commands

This approach gives you more control over the development process and helps ensure that Cline's actions align with your intentions.

Absolutely! Cline is an excellent learning companion. You can ask it to explain concepts, generate example code, or help you understand existing codebases.

Try prompts like "Explain how React hooks work" or "Show me a simple example of using async/await in JavaScript" to leverage Cline as a personalized coding tutor.

Yes! Cline can be a valuable code review assistant. Ask it to review specific files or pull requests, and it can identify potential bugs, suggest optimizations, point out security vulnerabilities, and recommend best practices.

While it shouldn't replace human code reviews, it can provide an additional layer of quality assurance.

Absolutely. Cline uses efficient strategies to handle large codebases, including partial-file analysis, summarization, and user-specified context.

Enterprise users particularly appreciate Cline's permission-based approach to file changes and command execution, ensuring security and control over the development process.

Yes. Cline was designed with enterprise needs in mind:

  • Permission gating on file writes and commands
  • Support for various AI providers to meet your security requirements
  • Transparent operation with clear approval workflows
  • Customizable rules to enforce your organization's coding standards

Cline works great for both serious enterprise development and casual coding sessions. You can ask it to quickly prototype ideas, refactor on the fly, or provide design suggestions—without a rigid, step-by-step process.

Cline adapts to your workflow, whether you're building a production application or experimenting with new concepts.

Prompting Cline

Custom instructions are Cline's baseline programming - they influence all interactions.

Setup steps:

  1. Open VSCode
  2. Click Cline extension settings
  3. Find "Custom Instructions" field
  4. Add your instructions

Common uses:

  • Enforce coding standards
  • Set quality requirements
  • Define error handling preferences
  • Establish documentation practices

.clinerules are project-specific instructions in your root directory:

  1. Create .clinerules in your project root
  2. Add project-specific guidelines:
    • Documentation requirements
    • Code standards
    • Testing rules
  3. Cline will automatically follow these rules

Cline adapts to your preferred coding style through:

  • Custom Instructions: Set global preferences for how Cline should write and format code
  • .clinerules: Define project-specific guidelines that Cline will follow
  • Learning from context: Cline analyzes your existing code to match its style and patterns

This ensures that Cline's contributions remain consistent with your codebase's standards.

Troubleshooting

With Claude 3.5 Sonnet:

  • You may hit concurrent connection limits
  • Add more credits through your Cline account at app.cline.bot
  • Break large tasks into smaller chunks

With free models (Gemini):

  • Expect frequent rate limits
  • Consider paid options for production
  • Try DeepSeek for cost-effective alternative

Pro tip: "Slow is smooth, and smooth is fast"

Quick fixes:

  • Check files are in your workspace root folder
  • Try closing and reopening the file

We value your feedback! You can:

  • Join our Discord community to share your experiences
  • Submit feature requests or bug reports on our GitHub repository
  • Reach out to our support team through the website

Your input helps us continuously improve Cline to better meet developers' needs.

Model Context Protocol (MCP)

MCP servers are like plugins that extend Cline's powers, letting it:

  • Connect to web services (like GitHub)
  • Control browsers
  • Access databases
  • Automate tasks
  • Handle project management
  1. Find a server:
  2. Add with Cline:
    • Just say: "Cline, add the MCP server from [GitHub URL]"
    • Cline handles the rest (cloning, building, config)

We welcome community contributions! Check out our GitHub repository to submit issues, suggest features, or open pull requests.

There's also an active community on Discord if you want to share feedback or help others.