utcp
All-in-one MCP server that can connect your AI agents to any native endpoint, powered by UTCP
claude mcp add --transport stdio universal-tool-calling-protocol-utcp-mcp npx @utcp/mcp-bridge \ --env UTCP_CONFIG_FILE="/path/to/your/.utcp_config.json"
How to use
UTCP-MCP Bridge exposes the Universal Tool Calling Protocol (UTCP) tools inside your MCP client without a traditional install. To get started, add the UTCP bridge to your MCP configuration as a server named utcp. The bridge runs via npx and loads your UTCP configuration from the UTCP_CONFIG_FILE you specify. Once running, it automatically registers all your UTCP manuals as MCP tools, giving you a unified interface to manage, search, and execute UTCP tools from Claude Desktop or other MCP clients. You can then call tools like register_manual, deregister_manual, call_tool, search_tools, and list_tools through the MCP interface just as you would with native MCP tools, but backed by UTCP-managed APIs and workflows. This setup enables seamless cross-tool orchestration across your UTCP ecosystem within MCP clients.
In practice, you configure UTCP with a .utcp_config.json file that describes your tools, how to load environment variables, and how results are post-processed. The bridge loads that configuration at startup, registers tools automatically, and presents a consistent MCP surface to manage and execute UTCP-enabled APIs and CLIs. If you already use UTCP tools, this bridge provides a drop-in way to expose them in your MCP client without further installation steps. The web UI is also available for advanced management and visibility into registered tools.
How to install
Prerequisites:
- Node.js and npm (for running via npx)
- Basic familiarity with UTCP configuration (.utcp_config.json)
Step-by-step:
-
Install Node.js and npm if you don’t already have them. You can download from https://nodejs.org/
-
Create or choose your UTCP config file at the path you’ll reference in UTCP_CONFIG_FILE (for example: /path/to/your/.utcp_config.json).
-
Add the MCP server configuration to your MCP client configuration (example shown in the Quick Start). This uses npx to fetch and run the bridge on demand:
{ "mcpServers": { "utcp": { "command": "npx", "args": ["@utcp/mcp-bridge"], "env": { "UTCP_CONFIG_FILE": "/path/to/your/.utcp_config.json" } } } }
-
Save your MCP client configuration and restart the MCP client or reload the configuration if supported.
-
Ensure the UTCP configuration file path is correct and the file is accessible by the runtime environment. The bridge will load this file on startup and register UTCP tools automatically.
-
(Optional) Open the web UI for UTCP bridge management if you plan to use web-based tooling: visit the web_ui_utcp_mcp_bridge directory as indicated in the repository documentation.
Notes:
- Because this is npx-based, there is no separate installation required for the bridge itself; it downloads and runs on demand.
- If you modify .utcp_config.json, some changes may require the bridge to reload tools or restart depending on how your MCP client handles dynamic updates.
Additional notes
Tips and common issues:
- Ensure UTCP_CONFIG_FILE points to a valid JSON configuration file and that the MCP client process has read access to it.
- If tools do not appear after startup, verify that the UTCP bridge successfully downloads (@utcp/mcp-bridge) and that the UTCP config file parses correctly.
- The bridge automatically registers all UTCP manuals as MCP tools; use list_tools to view registered tool names and tool_info to inspect individual tool schemas.
- You can load environment variables per the UTCP config; define them in UTCP_CONFIG_FILE to customize tool behavior per environment.
- For environments behind proxies, ensure Node.js networking is allowed to fetch the UTCP bridge npm package via npx.
- If you need to update UTCP tooling, update your .utcp_config.json and reload the bridge via your MCP client without reinstalling.
Related MCP Servers
mcp-agent
Build effective agents using Model Context Protocol and simple workflow patterns
nerve
The Simple Agent Development Kit.
python-utcp
Official python implementation of UTCP. UTCP is an open standard that lets AI agents call any API directly, without extra middleware.
prometheus
A Model Context Protocol (MCP) server that enables AI agents and LLMs to query and analyze Prometheus metrics through standardized interfaces.
mcp -odoo
A Model Context Protocol (MCP) server that enables AI assistants to securely interact with Odoo ERP systems through standardized resources and tools for data retrieval and manipulation.
neurolink
Universal AI Development Platform with MCP server integration, multi-provider support, and professional CLI. Build, test, and deploy AI applications with multiple ai providers.