$ loading_
Lets AI agents use live VS Code context for smarter code editing.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "vscode-agent-bridge" yet — see the docs or source repo.
Read the active file state and diagnostics from my current VS Code editor, then list the errors or warnings and suggest fixes by priority.
A live diagnostic summary of the current file with recommended fixes.
Using VS Code symbol navigation, find this function’s definition, references, and related classes, then summarize its role in the project.
Locations of relevant symbols, reference relationships, and a brief structural explanation.
Using the current editor context, perform a language-aware refactor on this file: rename unclear variables while preserving semantics.
Safer edits that stay consistent with the code context using editor-aware understanding.
While coding in VS Code, developers can let an AI read live diagnostics instead of relying only on files on disk, making it faster to find and fix issues. This is especially useful for unsaved problems already surfaced by the editor.
In complex projects, the AI can use symbol navigation to find definitions, references, and related code locations, helping users understand logic faster. It fits unfamiliar codebases or impact analysis of key functions.
Instead of editing based only on file text, this tool lets AI use language-aware information already known by the editor. It is useful for refactoring, targeted changes, and context-aware code edits.
It exposes live VS Code editor state, symbol navigation, diagnostics, and language-aware editing to AI agents through MCP. Its main purpose is to bridge the gap between what is on disk and what the editor already knows.
The difference is that it does more than read files on disk; it can also use the live editor state and language-service information from VS Code. This gives the AI context that is closer to the real development environment.
The provided materials do not include installation steps, runtime requirements, or key information. Please see the source repository for details.
Let AI agents debug code through VS Code in real sessions.
Let AI agents control the VS Code debugger across supported languages.
Let AI coding assistants reach restricted hosts through a local proxy automatically.
Give your VS Code agent real debugging with breakpoints, stepping, and inspection.
Connect VSCode to MCP so AI can inspect code context in real time.
Use Claude-style agents in Visual Studio for coding, debugging, testing, and navigation.