mcp-agent-to-vsc
Local MCP server that connects VS Code agents like Cline to OpenAI and provides workspace file tools (read, write, list) with restricted file access.
README
MCP - Agent to VSC
Local MCP bridge that connects an MCP-capable VS Code agent such as Cline to a workspace sandbox and an AI provider.
Architecture
Tauri / Rust desktop shell
|
v
TypeScript + Vite UI
|
+--> bundled MCP runtime (Node packaged executable)
| |
| +--> local gateway :3001
| +--> MCP stdio server
|
+--> Streamable HTTP MCP /mcp
|
+--> MCP server factory
+--> OpenAI API
Cline can connect directly to the stdio server or to the local Streamable HTTP endpoint.
MCP tools
ask_ai— send prompts to the configured AI modelread_file— read workspace fileswrite_file— write workspace fileslist_files— list workspace files/directoriesworkspace_info— inspect configuration
All file tools are sandboxed to WORKSPACE_ROOT and reject paths outside it.
Install
Requirements for development: Node.js 20+ and Rust for Tauri builds.
npm install
Copy .env.example to .env and set OPENAI_API_KEY if you want ask_ai enabled.
Run the local gateway
npm run gateway
The gateway exposes:
GET /api/healthGET /api/toolsGET /api/configPOST /api/promptPOST /mcp— Streamable HTTP MCP transport
The default MCP endpoint is http://127.0.0.1:3001/mcp.
Connect Cline automatically
The project includes a safe installer that preserves existing Cline MCP servers and creates a backup before changing the file:
npm run cline:install
For Streamable HTTP:
npm run gateway
npm run cline:install:http
For Cline CLI configuration:
npm run cline:install:cli
Set CLINE_MCP_SETTINGS_PATH to override automatic config discovery. The HTTP installer uses MCP_GATEWAY_URL when set.
Manual Cline configuration
For the stdio server:
{
"mcpServers": {
"mcp-agent-to-vsc": {
"command": "node",
"args": ["C:\\path\\to\\mcp-agent-to-vsc\\server.mjs"],
"disabled": false,
"autoApprove": []
}
}
}
For Streamable HTTP:
{
"mcpServers": {
"mcp-agent-to-vsc": {
"type": "streamableHttp",
"url": "http://127.0.0.1:3001/mcp",
"disabled": false,
"autoApprove": []
}
}
}
Frontend
npm run dev
Build the frontend with:
npm run build
Desktop + packaged runtime
The desktop shell uses Tauri/Rust while the UI uses TypeScript, Vite, HTML, and CSS. The Node gateway/MCP runtime is packaged into a platform-specific executable before Tauri bundles the application.
npm run runtime:build
npm run tauri:dev
npm run tauri:build
On Windows, the runtime is generated as src-tauri/binaries/mcp-agent-runtime-x86_64-pc-windows-msvc.exe and Tauri bundles it into the application. The desktop Rust command starts the bundled gateway automatically when the app launches.
The GitHub Actions Windows job also packages the runtime and uploads the generated NSIS .exe and MSI installers as workflow artifacts.
Development stack
- Frontend: TypeScript + Vite + HTML/CSS
- Gateway/MCP: Node.js + MCP SDK
- Desktop: Rust + Tauri
- Runtime packaging:
@yao-pkg/pkg - CI: GitHub Actions
The layers are intentionally separated; the project does not use one programming language for the entire application.
Security
- Never commit
.envor real API keys. - File operations are confined to
WORKSPACE_ROOT. - The Cline installer backs up an existing config before writing.
- Keep
GATEWAY_HOSTbound to127.0.0.1unless remote access is intentionally required.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.