Wanwu
China Unicom's Yuanjing Wanwu Agent Platform is an enterprise-grade, multi-tenant AI agent development platform. It helps users build applications such as intelligent agents, workflows, and rag, and also supports model management. The platform featur
What is Wanwu?
Wanwu is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to china unicom's yuanjing wanwu agent platform is an enterprise-grade, multi-tenant ai agent development platform. it helps users build applications such as intelligent agents, workflows, and rag, and a...
China Unicom's Yuanjing Wanwu Agent Platform is an enterprise-grade, multi-tenant AI agent development platform. It helps users build applications such as intelligent agents, workflows, and rag, and also supports model management. The platform featur
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- China Unicom's Yuanjing Wanwu Agent Platform is an enterpris
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx wanwuConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Wanwu
Wanwu (万物) is China Unicom's Yuanjing enterprise-grade, multi-tenant AI agent development platform that lets teams build intelligent agents, low-code workflows, and RAG-based knowledge bases from a visual interface. It integrates over 100 pre-selected MCP servers, supports more than a dozen document formats for retrieval-augmented generation, and provides a model hub for managing proprietary and open-source LLMs in one place. Enterprises use it to deploy customer-facing agents, automate internal workflows, and connect AI reasoning to CRM, ERP, and OA systems via its REST API without writing infrastructure code.
Prerequisites
- Docker and Docker Compose installed on the host machine
- Hardware: 8-16 CPU cores, 32 GB RAM, 200 GB storage minimum
- An LLM with a context window of at least 32,000 tokens (e.g., Qwen, GPT-4, Claude)
- Network access to download Docker images from the registry
Clone the repository
Clone the Wanwu source repository to obtain the Docker Compose configuration and environment template.
git clone https://github.com/UnicomAI/wanwu.git
cd wanwuConfigure environment variables
Copy the example environment file and edit the required variables for your deployment.
cp .env.example .envEdit the .env file with your settings
Set your system architecture, external IP, and a strong random JWT signing key. The external IP can be 'localhost' for single-machine deployments.
# In .env:
WANWU_ARCH=amd64
WANWU_EXTERNAL_IP=localhost
WANWU_BFF_JWT_SIGNING_KEY=your-random-complex-string-hereCreate the Docker network and start services
Create the required Docker network, then bring up all Wanwu services using Docker Compose.
docker network create wanwu-net
docker compose --env-file .env --env-file .env.image.amd64 up -dAccess the platform and change the default password
Open the Wanwu web UI, log in with the default credentials, and immediately change the admin password.
# Open in browser:
http://localhost:8081
# Default credentials:
# Username: admin
# Password: Wanwu123456Add a model and create your first agent
Navigate to Model Hub, add your LLM provider API key, then go to Agent to create a new intelligent agent using the visual builder.
Wanwu Examples
Client configuration
To connect to Wanwu's MCP endpoint from an external MCP client, use the HTTP/SSE transport URL exposed by the platform.
{
"mcpServers": {
"wanwu": {
"type": "sse",
"url": "http://localhost:8081/mcp/sse"
}
}
}Prompts to try
Example prompts to use once connected to a Wanwu agent.
- "Search our internal knowledge base for the Q3 product roadmap and summarize the key milestones."
- "Create a new workflow that sends a daily Slack summary of open support tickets."
- "Use the GraphRAG knowledge base to find all documents related to data privacy compliance."
- "List all available MCP tools and show me which ones relate to CRM integration."Troubleshooting Wanwu
Docker Compose fails to start with network errors
Ensure the wanwu-net Docker network exists before starting ('docker network create wanwu-net'). If services still fail, check that required ports (8081, etc.) are not occupied by other processes.
LLM responses are truncated or the agent fails mid-task
Wanwu requires models with at least a 32,000-token context window for agentic tasks. Switch to a model that meets this requirement in the Model Hub settings.
RAG retrieval returns irrelevant results
Ensure documents have been fully indexed after upload (check the Knowledge Base status indicator). GraphRAG indexing can take several minutes for large document sets; wait for the status to show 'Ready' before querying.
Frequently Asked Questions about Wanwu
What is Wanwu?
Wanwu is a Model Context Protocol (MCP) server that china unicom's yuanjing wanwu agent platform is an enterprise-grade, multi-tenant ai agent development platform. it helps users build applications such as intelligent agents, workflows, and rag, and also supports model management. the platform featur It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Wanwu?
Follow the installation instructions on the Wanwu GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Wanwu?
Wanwu works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Wanwu free to use?
Yes, Wanwu is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Wanwu Alternatives — Similar Coding Agents Servers
Looking for alternatives to Wanwu? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Wanwu in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Wanwu?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.