BifrostMCP

AI coding tools have a fundamental information problem. They read text files, infer structure from symbols and naming, and make educated guesses about what calls what and where things are defined. VS Code already knows all of this exactly. Every time you hit go-to-definition or find-all-references, the language server answers in milliseconds with precise, type-checked answers. That information just was not accessible to AI tools.

BifrostMCP is a VS Code extension that acts as an MCP server, bridging that gap. It exposes language server features directly to any AI tool that supports the Model Context Protocol: go-to-definition, find-all-references, get-hover-info, rename-symbol, and about a dozen others. Instead of an AI guessing where a function is defined or what implements an interface, it can ask VS Code and get the exact answer.

Benefits to models

This approach has major benefits to models, primarily in terms of correctness and saving context. Current AI tooling will use grep when it wants to find a symbol. This is inneficient because grep returns all occurences of the token. Even partial matches (IsValid within IsValidEmail), function calls for other type heirarchies (phone.IsValid instead of email.IsValid), and usages of a symbol instead of the declaration.

By allowing models to more accurately express their intent "Search for the declaration of this symbol", "Give me all usages of this symbol", they are able to get more accurate results which improves performances and saves context.

It also allows models to make changes faster. For example, using the rename tool a model can edit a symbol that may be used 10s or 100s of times. Doing so with grep and file edits would require significantly more tokens and tool calls.

Adoption without marketing

I posted it in a few relevant communities and that was it. 205 GitHub stars and 1,600+ VS Code extension installs came from developers discovering it organically. Developers using Claude, Cursor, and other MCP-compatible tools have picked it up because it solves a real problem they hit immediately when trying to use AI tools on large codebases. When a tool genuinely fills a gap, it finds its users.