openmcp-client
All in one vscode plugin for mcp developer
claude mcp add --transport stdio lstm-kirigaya-openmcp-client npx -y openmcp-client \ --env OPENMCP_CLIENT_CONFIG="Path to mcp configuration or environment-specific settings (if required)"
How to use
OpenMCP OpenMCP Client is an integrated workflow designed for MCP development and testing. The openmcp-client module provides an Integrated Inspector and MCP client basic functions, allowing you to test MCP tools, prompts, and resources directly within a single environment. You can place tested tools into the Interactive Testing module to stress-test large model interactions, and then use the project-level management panel to organize work at both project and global levels. It supports multiple large models, XML mode, and customizable tool selection, making it suitable for debugging and validating MCP configurations before deploying agents with the OpenMCP SDK.
To begin, install or run the openmcp-client in your development environment, then open the inspector to browse available MCP tools and test prompts. Use the inspector to load a specific MCP configuration, pick tools you want to test (for example, web-search, summarize, or data extraction tools), and execute prompts to observe how the MCP server orchestrates tool usage and model responses. Once you have a validated configuration, you can generate or load a debugged mcpconfig.json into the SDK workflow to deploy an agent that can perform tasks end-to-end against real data.
How to install
Prerequisites:
- Node.js v14 or later
- npm (comes with Node.js) or yarn
Installation steps:
-
Ensure Node.js and npm are installed:
- node -v
- npm -v
-
Install the MCP client package (openmcp-client) locally or run via npx for quick start:
-
If using npx (no global install): npm check-version >/dev/null 2>&1 || true npx -y openmcp-client
-
If you prefer a local install (recommended for development): mkdir openmcp-project cd openmcp-project npm init -y npm install openmcp-client --save
Run via node script or npm script depending on your setup
-
-
If your project requires the OpenMCP SDK for agent deployment, install it as well: npm install openmcp-sdk
-
Create or place your mcpconfig.json in your project root, or use the inspector to generate it after testing with the client.
-
Start the client server or development environment as appropriate for your setup (see mcp_config for invocation details).
Additional notes
Tips and common issues:
- If you encounter missing tool support, verify that the tools you expect are listed in the Interactive Testing module and that XML mode options are enabled if you rely on XML tool configurations.
- Ensure your Node.js version is compatible with openmcp-client and any dependencies used by the inspector or tools.
- When using the npx invocation, the package is fetched on demand; for a stable development environment, prefer a local install and run the client via npm scripts.
- The environment variable OPENMCP_CLIENT_CONFIG can be used to point the client to a specific mcpconfig.json or to toggle environment-specific settings; adjust as needed for your deployment.
- If you plan to deploy agents with the OpenMCP SDK, ensure the agent’s runtime environment has network access to any tools or APIs your MCP uses.
- Regularly back up your mcpconfig.json and project-level settings to avoid losing customized tool configurations during updates.
Related MCP Servers
dexto
A coding agent and general agent harness for building and orchestrating agentic applications.
aser
Aser is a lightweight, self-assembling AI Agent frame.
joinly
Make your meetings accessible to AI Agents
moling
MoLing is a computer-use and browser-use based MCP server. It is a locally deployed, dependency-free office AI assistant.
omega-memory
Persistent memory for AI coding agents
serper
A Serper MCP Server