v3.3.0

• You can now create a .clineignore file to block Cline from accessing specified file patterns. Especially useful when using auto-approve in a project with sensitive files!
• Adds keyboard shortcut (CMD + Shift + A) to easily switch between Plan/Act mode
• Fixes the infamously annoying bug where new files wouldn't show up in textfield files dropdown list!
• Adds automatic intelligent retry for rate limited requests (thanks @ViezeVingertjes )
• Adds support for AWS Bedrock profiles + Requesty, Together, Alibaba Qwen providers

v3.2.9

Full Changelog: v3.2.8...v3.2.9

v3.2.6

• Save last used API/model when switching between Plan and Act, for users that like to use different models for each mode
• New Context Window progress bar in the task header to understand increased cost/generation degradation as the context increases, and be aware when the older messages are removed to stay within window
• Add Advanced Settings to remove MCP prompts from requests to save tokens, enable/disable checkpoints for users that don't use git (more coming soon!)

v3.2.0

• New Plan/Act mode toggle! Plan mode turns Cline into an architect that gathers information, asks clarifying questions, and designs a solution. Switch back to Act mode to let him execute the plan! See a demo here: https://x.com/sdrzn/status/1881761978986934582

• Easily switch between API providers and models using a new popup menu under the chat field

• Add VS Code LM API provider to run models provided by other VS Code extensions (e.g. GitHub Copilot). Shoutout to @julesmons and @RaySinner for putting this together!

• Add on/off toggle for MCP servers to disable them when not in use, and auto-approve setting for individual tools in MCP servers. Thanks @mrubens!

v3.1.0

Checkpoints: Compare and Restore

Whenever Cline uses a tool, the extension takes a snapshot of the workspace so you can easily restore back to that point. When you hover over messages, you’ll notice two new buttons:

• Compare lets you see a diff overview between the snapshot and the current state of your workspace
• Restore lets you choose one of three options. While you’ll mostly use the ‘Restore Task and Workspace’ option, ‘Restore Task Only’ can be useful for when you need to course correct or undo a problematic tool use, and ‘Restore Workspace Only’ can be useful to checkout different versions of your code without losing Cline’s work.

See new changes

There is now a ‘See new changes’ button when Cline completes a task, giving you an overview of all the changes made to your workspace (since the last 'Task Completed'). This keeps you in the loop with Auto-approve enabled, saving you the trouble of reviewing a task’s message history.

Task Size

Under the hood, the extension uses git to efficiently track changes, which uses additional disk space when starting new tasks in Cline. Keep track of how much space a task takes with the new delete button, allowing you to manage disk usage easily.

v3.0.0

Auto-approve

Cline's most requested feature is here 🎉 You can now let Cline work more autonomously with the new Auto-approve menu, letting you choose what tools he can use without needing your permission.

• Enable or disable for: reading files & directories, editing files, executing terminal commands, using the browser, and using MCP servers.
• Set a limit for how many API requests Cline makes before asking for your approval to keep going.
• And if you have Cline working in the background, he can send system notifications for when he needs your attention, i.e. to get your approval, answer a question, or when a task is completed.

Diff editing for large files

When Cline needs to make targeted edits to large files, he now uses a search & replace diff format where he only has to output the content to replace, rather than the whole file. This allows for faster, more reliable edits and prevents code deletion issues where previously, unmodified sections of code might have been replaced with comments like // rest of code here. Cline can also fall back to whole file editing for when it makes sense, e.g. replacing boilerplate files, complete refactors, or when diff editing fails.

.clinerules

You can now specify custom instructions for your project by adding a root-level .clinerules file. This allows you to define project-specific behaviors and guidelines for Cline–perfect for setting conventions, pointing to important documentation, or providing context about your project's architecture.

v2.2.0

Add custom tools to Cline using MCP!

The Model Context Protocol allows agents like Cline to plug and play custom tools, e.g. a web-search tool or GitHub tool. You can now add and configure MCP servers by clicking the new server icon in the menu bar.

To take things a step further, Cline also has the ability to create custom tools for himself. Just say "add a tool that..." and watch as he builds and installs new capabilities specific to your workflow. For example:

• "...fetches Jira tickets": Get ticket ACs and put Cline to work
• "...manages AWS EC2s": Check server metrics and scale up or down
• "...pulls PagerDuty incidents": Pulls details to help Cline fix bugs

Cline handles everything from creating the MCP server to installing it in the extension, ready to use in future tasks. The servers are saved to ~/Documents/Cline/MCP so you can easily share them with others too.

Try it yourself by asking Cline to "add a tool that gets the latest npm docs", or see a demo of MCP in action here.

v2.1.0

Cline now uses Anthropic's new "Computer Use" feature to launch a browser, click, type, and scroll. This gives him more autonomy in runtime debugging, end-to-end testing, and even general web use. Try asking "look up the weather in Colorado" to see it in action! (Available with Claude 3.5 Sonnet v2)

See a demo here! https://x.com/sdrzn/status/1850880547825823989

v2.0.0

• New name: Meet Cline, an AI assistant that can use your CLI aNd Editor. While “Claude Dev” was a tribute to Claude 3.5 Sonnet, v2.0 brings updates that significantly improve performance across other models making it a good opportunity to give the extension its own identity.

• No more staring at loading spinners: Responses are now streamed into the editor + a yellow text decoration animation to keep track of Cline’s progress as he edits files. Watching Cline work feels like magic and I’m so incredibly excited for you all to experience it.
  

• Cancel button: With the new streaming feature, you can now Cancel mid-generation to give Cline feedback if he goes off in the wrong direction. This gives you more control over tasks, leading to the results you want faster.

• 40% fewer requests to accomplish tasks + better performance with other models: This update replaces the rigid tool use API with a customized XML tags approach, allowing for more flexible tool interactions and significantly improving results for non-Claude models. Behind the scenes, older versions of the extension used an API feature called “Tool Use” (AKA “Function Calling”) where tools like read_file and execute_command were defined in a strict JSON schema sent with each request, requiring models to respond with a structured output that had to fit into this schema. While some models like Claude 3.5 Sonnet are heavily fine-tuned to be good at this, other models like Llama and DeepSeek are not as adept, often leading to inability to complete even the most basic tasks. This not only limited the extension’s accessibility with other models, but also diminished Claude 3.5 Sonnet’s potential as forcing any model to respond in structured output has been proven to have negative qualitative effects. To eliminate these bottlenecks while still giving models the ability to invoke tools, I took inspiration from this study showing how prompt-engineering can enable stable tool-calling capabilities + Anthropic’s prompt to achieve their Artifacts feature (uses tags like <antArtifact>) to create a completely re-imagined system prompt where models are effectively fine-tuned on the spot how to best interact with Cline’s toolset. Instead of having to respond in a rigid JSON format, models can now invoke tools in unstructured text with XML tags like <read_file> and <execute_command>. Since we define the tool use guidelines ourselves we can even tailor it with context like “these tools are executed upon the user's approval” instead of relying on the generalized internal prompts that Anthropic uses to achieve tool calling for their API. To quantify the improvement these changes made, I ran a simple benchmark: I gave the following task in a Next.js app “Implement a dark mode toggle feature, and save the user’s preference to persist across sessions” and measured the # of requests until the task was successfully completed. This was a practical scenario that tested models’ abilities to use the appropriate tools to explore an existing project’s source code, apply edits, and debug issues that came up along the way.
  
  Claude 3.5 Sonnet saw a ~40% reduction in # of requests required to accomplish the task, while other models like Llama and DeepSeek are now not only able to tool call more reliably but seem to have a greater capacity to reason about executing tasks successfully.
  While the native tool calling API was a core feature of earlier versions of the extension, it’s become clear that it limited the models' performance. The more flexibility the models have in generating responses, the better the outcomes. Although it’s more work on the extension’s part to parse the XML tool invocations, it is certainly worth the trouble.

• Search and use any OpenRouter model: the extension automatically fetches the latest model list + pricing details, giving you access to the newest models as soon as they’re available. Try searching “free” to find no-cost options.

• Apache 2.0 License: Changed the project’s license from MIT to Apache 2.0 to provide stronger open-source protections for the future development of the project. Thank you contributors for making this a smooth process.

• New README and got a cool domain for the project https://cline.bot (wip). On a final note, I also wanted to share that I’m now working on Cline full-time and am excited to keep improving the extension and make it the best thing it can possibly be. Thank you all for the incredible support around the project.

–Saoud

v1.9.0

• Claude can now use a browser! This update adds a new inspect_site tool that captures screenshots and console logs from websites (including localhost), making it easier for Claude to troubleshoot runtime/visual issues on his own. See a demo here.
  
• Improved automatic linter/compiler debugging by only sending Claude new errors that result from his edits, rather than reporting all workspace problems.