BroPilot
BroPilot is a Visual Studio and SQL Server Management Studio extension that brings the power of local Large Language Models (LLMs) directly into your development workflow. Unlike cloud-based AI assistants, BroPilot enables intelligent code assistance—such as code completions, explanations, and refactoring suggestions—without sending your code or context to external servers. This ensures privacy, low latency, and full control over your development environment.
🚀 Install from VS Marketplace ✨
Problem Statement
Modern software development increasingly relies on AI-powered code assistants to boost productivity, improve code quality, and reduce repetitive tasks. However, most existing solutions require sending sensitive code to third-party cloud services, raising concerns about privacy, data security, and compliance. Additionally, reliance on external APIs can introduce latency, cost, and dependency risks.
BroPilot addresses these challenges by:
- Allowing developers to leverage advanced code assistance features powered by local LLMs.
- Ensuring that all code and context remain on the developer's machine.
- Providing a seamless, integrated experience within Visual Studio.
BroPilot is engineered for high performance and a smooth user experience, even when working with large codebases or complex LLM interactions.
- Responsive UI: The extension is designed to keep the Visual Studio interface fast and responsive at all times. Intensive operations are offloaded to background threads, ensuring that the UI never blocks or stutters.
- Streaming Token Processing: LLM responses are streamed in real time. As tokens are generated by the model, they are processed and displayed incrementally, allowing you to see results immediately rather than waiting for the entire response.
- Deferred Rendering: To further enhance performance, BroPilot defers rendering of expensive operations such as syntax highlighting. This means that only the necessary parts of the interface are updated as new data arrives, minimizing unnecessary redraws and improving perceived speed.
- Optimized for Large Projects: Special care has been taken to ensure that BroPilot remains efficient and responsive, even in solutions with thousands of files or very large code documents.
These optimizations collectively ensure that BroPilot feels fast, fluid, and unobtrusive, letting you focus on your code without interruption.
This method is analyzed in a 11mb file
Simple interactions are really fast
Key Features
1. Local LLM Integration
- Connects to local LLM servers (such as Ollama, LM Studio, or other compatible backends).
- No code or context is sent to the cloud—everything stays on your machine.
2. Code Explanations
- Select code and receive clear, concise explanations powered by your local model.
- Useful for onboarding, code reviews, and understanding legacy code.
3. Refactoring Assistance
- Get suggestions for improving code structure, readability, and maintainability.
- Supports common refactoring patterns and best practices.
4. Seamless Visual Studio Integration
- Native Visual Studio extension—no need to switch tools or editors.
5. Configurable and Extensible
- Easily configure the extension to connect to your preferred local LLM backend.
6. Privacy and Security
- All processing is performed locally; no data leaves your machine.
- Ideal for organizations with strict compliance or confidentiality requirements.
This feature enables users to apply code blocks directly within the current file in a sophisticated, embedded way. It leverages Visual Studio’s internal APIs for safe and context-aware code insertion.
Key Capabilities:
- Add class or method: Generate and insert new classes or methods into the current C# file.
- Replace class or method: Replace existing code with suggested implementations using intelligent generation.
Limitations:
- Language Support: Currently limited to C# files.
- Solution Context: The feature works only when the file is part of an active Visual Studio solution.
When BroPilot Try tries to apply the changes, it will output its decision and motivation.
SQL Server Management Studio (SSMS) Support
BroPilot is no longer limited to Visual Studio—developers and database professionals can now benefit from its local AI assistance directly inside SQL Server Management Studio. This integration makes it easier than ever to understand, write, and refactor SQL queries with the same privacy-first approach.
Key Capabilities in SSMS:
Query Explanations – Select SQL statements and instantly get clear explanations of what the query does.
Query Suggestions & Refactoring – Improve readability, performance, or maintainability of SQL queries with locally powered AI suggestions.
Seamless Workflow – BroPilot integrates naturally into SSMS, so you can work with your databases without switching tools.
Same Privacy, Same Performance – Just like in Visual Studio, all processing happens locally. No queries or schema data are sent to external servers.
This makes BroPilot a valuable companion not only for software developers, but also for database administrators, analysts, and anyone working extensively with SQL Server.
Getting Started
Install BroPilot
Download and install the BroPilot extension from the Visual Studio Marketplace or build from source.
Set Up a Local LLM Server
Start your preferred local LLM backend (e.g., Ollama, LM Studio) and ensure it is accessible.
Configure BroPilot
In Visual Studio, open the BroPilot settings and specify the connection details for your local LLM server.
Start Coding
Use BroPilot features directly from the editor—trigger completions, request explanations, or refactor code with a click.
Requirements
- Visual Studio 2022 or later
- .NET Framework 4.8 and/or .NET 8 (for project compatibility)
- A running local LLM server (Ollama, LM Studio, or compatible)
- Lowest GPU I tried: RTX2070 Super 8gb (used Qwen3-4b)
License
BroPilot is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
This means you are free to use and share the extension for non-commercial purposes, but you may not modify or use it for commercial gain.
Acknowledgments
- Inspired by the growing ecosystem of local LLMs and privacy-first developer tools.
- Thanks to the open-source community for making local AI accessible to all.
For support, questions, or feedback, please open an issue on the GitHub repository.
