DiagBridge MCP
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@DiagBridge MCPdiagnose my Windows computer's network issues"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
DiagBridge MCP
DiagBridge MCP is a general Windows MCP bridge for trusted local and remote AI agents.
It exposes one shared tool registry through:
stdiofor local MCP clients (Codex, Claude Code, Cursor, etc.)Streamable HTTPfor remote MCP clients (ChatGPT, Codex Plugin, remote agents, etc.)
Runtime Requirements: Requires Node.js 22.6 or newer. Tested on Node.js 24.
中文:
DiagBridge MCP 是面向可信本地和远程 AI Agent 的通用 Windows MCP 桥。 它通过 stdio 和 Streamable HTTP 暴露同一套工具。
Transports
Entry point | Command / Script | Purpose |
Stdio MCP server (MCP Host) |
| Production integration for MCP hosts over stdio. |
Stdio MCP server (Manual Dev) |
| Manual process launch for development. |
| Streamable HTTP MCP transport | npm run dev:remote-mcp | Remote MCP client integration over Streamable HTTP (src/mcp/transports/streamable-http.ts). |
| Local Control Panel (Web UI) | npm run dev:friend-ui | Local Web UI (http://127.0.0.1:8790) and Remote Streamable HTTP server for friend diagnostic session management. |
Related MCP server: MCP Server for WinDbg Crash Analysis
Tools Overview
All tools are defined in a single, shared tool registry (src/mcp/tool-registry.ts).
Local Stdio Defaults (6 read-only tools)
system_infolist_dirread_filedrive_inventoryjunk_candidateswindows_event_summary
write_file and run_command are destructive and disabled by default. They can be enabled for stdio using DIAGBRIDGE_MCP_TOOLS.
Remote Streamable HTTP Defaults (4 read-only tools)
system_infodrive_inventoryjunk_candidateswindows_event_summary
read_file, write_file, and run_command are never registered on the remote transport for security and privacy protection.
For full tool details, see docs/TOOLS.md.
Configuration
Environment variables:
Variable | Default | Meaning |
|
| Default host address. |
|
| Bind address for Streamable HTTP server ( |
|
| Default port number. |
|
| Streamable HTTP server port ( |
| generated at startup | Session token for authenticating HTTP requests. |
| read-only local tools | Comma-separated enabled tools for local stdio transport ( |
|
| Enable |
| current directory | Base working directory. |
|
| JSONL audit log file path. |
Quick Start
npm install
npm run check
npm test
# Launch remote Streamable HTTP server
npm run dev:remote-mcpDocumentation
This server cannot be deployed
Maintenance
Related MCP Connectors
Read-only finance and operations controls for AI agents with evidence and safe next actions.
Turns a phone into a camera+Bluetooth remote so AI assistants can see and control any PC.
Provides capabilities that let LLM agents perform a range of infrastructure management tasks.
Trust signals for AI agents: an open agent-readiness standard and developer tool guide. Read-only.
Related MCP Servers
- FlicenseBqualityDmaintenanceProvides Windows system diagnostic capabilities to AI agents, allowing them to access event logs, crash information, system uptime, and perform stability analysis.113-
- AlicenseNot gradedqualityNot gradedmaintenanceBridges AI models with WinDbg to analyze Windows crash dumps and perform remote debugging through natural language queries, enabling execution of debugger commands and automated crash analysis.MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI assistants to diagnose, troubleshoot, and manage PC settings on Windows, macOS, and Linux through natural language commands.GPL 3.0
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to monitor system resources and control media, brightness, volume, and other Windows PC functions through natural language commands.1MIT