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 features a developer-friendly license, and we welcome all developers to build upon the platform.
claude mcp add --transport stdio unicomai-wanwu docker run -i unicomai/wanwu:latest \ --env WANWU_CONFIG="path/to/config (optional)" \ --env WANWU_LOG_LEVEL="info (optional: debug, info, warn, error)"
How to use
Wanwu is an enterprise-grade AI agent platform that provides a modular MCP (Multi-Component Protocol) interface allowing AI models to connect with external tools, knowledge bases, and workflows. The Wanwu MCP server exposes standardized endpoints and workflows to enable seamless tool integration, Web Search, knowledge base retrieval, and low-code workflow orchestration. With Wanwu, you can leverage pre-built MCP interfaces to connect common data sources and services, or extend the platform to your own business processes. Typical usage involves deploying Wanwu as a Docker container, configuring your MCP endpoints and data sources, and then interacting with the platform through its MCP-enabled agents and tooling to orchestrate end-to-end AI-assisted workflows.
How to install
Prerequisites:
- Docker installed on the host with access to pull images from the registry
- Adequate CPU/RAM resources for running MCP components (recommended: 2 CPU cores, 4GB RAM minimum)
- Network access to required services (databases, knowledge bases, web search sources) as needed by your configuration
Installation steps:
-
Pull and run the Wanwu MCP Docker image
- Ensure Docker is running on your machine
- If you want to keep the config external, mount a directory containing Wanwu configs
Example command: docker run -d --name wanwu-mcp
-p 8080:8080
-v /path/to/wanwu/configs:/app/configs
unicomai/wanwu:latest -
Configure environment variables (optional but recommended)
- WANWU_CONFIG: Path or identifier for your Wanwu configuration
- WANWU_LOG_LEVEL: log verbosity (info, debug, warn, error)
-
Verify the container is running docker ps docker logs wanwu-mcp
-
Access the MCP endpoints and begin integration
- Use the exposed API endpoints as documented in Wanwu MCP documentation (the exact URLs depend on your deployment and port mappings)
Notes:
- If you are customizing MCP integrations, prepare your external tools (knowledge bases, APIs, databases) and point Wanwu to them via the config directory mounted above.
Additional notes
Tips and common issues:
- Ensure the Docker image tag is latest or a trusted release tag compatible with your environment
- If you encounter network issues, verify firewall rules and that outbound access to required services is allowed
- For production deployments, consider mounting persistent storage for logs and configuration, and enable proper log rotation
- Environment variables are optional but recommended to tailor Wanwu behavior to your enterprise data policies (e.g., WANWU_LOG_LEVEL, WANWU_CONFIG)
- If MCP interfaces fail to load, check that all required dependencies (web search sources, MCP adapters, and knowledge bases) are accessible and correctly referenced in your config
Related MCP Servers
dify
Production-ready platform for agentic workflow development.
bytebot
Bytebot is a self-hosted AI desktop agent that automates computer tasks through natural language commands, operating within a containerized Linux desktop environment.
cursor-talk-to-figma
TalkToFigma: MCP integration between AI Agent (Cursor, Claude Code) and Figma, allowing Agentic AI to communicate with Figma for reading designs and modifying them programmatically.
ag2
AG2 (formerly AutoGen): The Open-Source AgentOS. Join us at: https://discord.gg/sNGSwQME3x
koog
Koog is the official Kotlin framework for building predictable, fault-tolerant and enterprise-ready AI agents across all platforms – from backend services to Android and iOS, JVM, and even in-browser environments. Koog is based on our AI products expertise and provides proven solutions for complex LLM and AI problems
paperdebugger
A Plugin-Based Multi-Agent System for In-Editor Academic Writing, Review, and Editing