BifrostMCP is a VSCode extension that acts as an MCP server, bridging the gap between AI coding assistants and VSCode's powerful development tools. It empowers AI tools with advanced code navigation, analysis, and manipulation capabilities, enhancing the AI's understanding and interaction with your codebase.

This MCP solution exposes VSCode's language server features, offering functionalities like finding usages, go-to-definition, symbol search, and code actions. By integrating with BifrostMCP, AI models can access semantic tokens, document symbols, and type information, enabling smarter code completion, refactoring suggestions, and more.

BifrostMCP seamlessly integrates with AI assistants via HTTP/SSE, adhering to the MCP protocol. Developers benefit from improved AI-driven code assistance, leading to increased productivity and code quality. Installation is straightforward, requiring only the VSCode extension and configuration within your AI assistant (like Cursor or Cline). It also supports multi-project setups with dedicated ports and endpoints.

Language Server Integration

BifrostMCP leverages VSCode's Language Server Protocol (LSP) integration to provide AI coding assistants with deep semantic understanding of code. This allows the AI to access language-specific features like syntax highlighting, code completion, and error checking, regardless of the programming language being used. By acting as a bridge between the AI and the language server, BifrostMCP enables more intelligent and accurate code analysis and manipulation. For example, an AI assistant can use the language server to identify all instances of a variable, understand its type, and suggest appropriate code changes. This integration supports any language for which a VSCode language extension exists.

Advanced Code Navigation

BifrostMCP exposes VSCode's code navigation capabilities to AI tools, enabling them to perform tasks such as finding all references to a symbol, jumping to its definition, and discovering implementations of an interface. This allows AI assistants to understand the relationships between different parts of the codebase and navigate complex projects with ease. For instance, when an AI is asked to refactor a function, it can use BifrostMCP to quickly identify all locations where the function is called and ensure that the changes are applied consistently throughout the project. This feature significantly enhances the AI's ability to perform complex code modifications and maintain code quality.

Workspace Symbol Search

BifrostMCP provides AI tools with the ability to search for symbols (variables, functions, classes, etc.) across the entire workspace. This allows the AI to quickly locate specific code elements, regardless of their location in the project. This is particularly useful for large projects with many files and directories. For example, an AI assistant can use this feature to find all classes that inherit from a specific base class, or to identify all functions that use a particular API. The get_workspace_symbols tool enables AI to understand the overall structure of the project and identify relevant code elements for a given task.

Multi-Project Support

BifrostMCP supports multiple projects by allowing each project to have its own dedicated MCP server endpoint and port. This is crucial when working with multiple VS Code windows or projects that require language server features. By creating a bifrost.config.json file in each project's root directory, developers can specify a unique port and path for each project. This ensures that each project has its own isolated environment and avoids conflicts between different projects. For example, a developer working on both a frontend and a backend project can configure BifrostMCP to use different ports for each project, allowing the AI assistant to seamlessly switch between the two projects without any interference.

HTTP/SSE Server

BifrostMCP uses a standard HTTP/SSE (Server-Sent Events) server to expose VSCode's language features to AI assistants. This allows AI tools to communicate with BifrostMCP using a simple and widely supported protocol. The SSE protocol provides a persistent connection between the AI and BifrostMCP, allowing for real-time updates and efficient communication. This is particularly important for interactive AI coding assistants that need to respond quickly to user input. The use of HTTP/SSE also makes it easy to integrate BifrostMCP with a variety of AI tools and platforms.