Ragent
企业级 Agentic RAG 智能体 - 全链路覆盖文档解析、多路检索、意图识别、问题重写、会话记忆、MCP 工具调用与深度思考。面向真实业务场景，从 0 到 1 完整工程实现。
What is Ragent?
Ragent is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to 企业级 agentic rag 智能体 - 全链路覆盖文档解析、多路检索、意图识别、问题重写、会话记忆、mcp 工具调用与深度思考。面向真实业务场景，从 0 到 1 完整工程实现。
企业级 Agentic RAG 智能体 - 全链路覆盖文档解析、多路检索、意图识别、问题重写、会话记忆、MCP 工具调用与深度思考。面向真实业务场景，从 0 到 1 完整工程实现。
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- 企业级 Agentic RAG 智能体 - 全链路覆盖文档解析、多路检索、意图识别、问题重写、会话记忆、MCP 工具调用
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx ragentConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Ragent
Ragent is an enterprise-grade Agentic RAG system built with Spring Boot 3 and Spring AI, providing a full-pipeline implementation that covers document parsing, multi-channel parallel retrieval, intent recognition, query rewriting, session memory management, MCP tool invocation, and deep reasoning. It is designed for real business scenarios with production features like three-state circuit breakers, distributed rate limiting, request trace propagation, and automatic model failover. Developers use Ragent as a reference implementation or foundation for building production RAG-based AI agents that go beyond simple question-answering.
Prerequisites
- Java 21 or higher (JDK 21+) with Maven 3.8+
- A supported LLM API key (OpenAI-compatible endpoint or configured model provider)
- A vector database for embeddings (the project integrates with Spring AI's vector store abstraction)
- Redis for distributed rate limiting, caching, and session management
- PostgreSQL or compatible relational database for conversation and document metadata
Clone the repository
Clone the Ragent repository from GitHub and navigate into the project directory.
git clone https://github.com/nageoffer/ragent.git
cd ragentConfigure infrastructure and AI provider
Edit the application configuration to set your database connection, Redis, vector store, and LLM API credentials. The project uses a Maven multi-module structure with framework, infra-ai, and bootstrap layers.
# Edit bootstrap/src/main/resources/application.yml
# Key settings:
# spring.datasource.url, username, password
# spring.redis.host, port
# spring.ai.openai.api-key (or equivalent)
# spring.ai.vectorstore configurationBuild the project
Build all Maven modules from the project root. The multi-module structure requires building from the top level.
mvn clean install -DskipTestsRun the application
Start the Spring Boot bootstrap module to launch the full Ragent server including the RAG pipeline, MCP tool integration, and REST/SSE APIs.
mvn spring-boot:run -pl bootstrapIngest documents into the knowledge base
Upload documents through the REST API. The ingestion pipeline will parse, chunk, embed, and store them in the vector database for retrieval.
curl -X POST http://localhost:8080/api/document/upload \
-H 'Content-Type: multipart/form-data' \
-F '[email protected]'Configure MCP tool integration
Ragent supports MCP tool calling for extending agent capabilities with external services. Configure MCP server connections in the application properties to enable business tool invocation during RAG conversations.
{
"mcpServers": {
"ragent": {
"command": "npx",
"args": ["ragent"]
}
}
}Ragent Examples
Client configuration
MCP client configuration for connecting to the Ragent server.
{
"mcpServers": {
"ragent": {
"command": "npx",
"args": ["ragent"]
}
}
}Prompts to try
Example prompts that exercise Ragent's multi-channel retrieval, intent recognition, and session memory capabilities.
- "Based on the uploaded documentation, what is the process for onboarding a new enterprise customer?"
- "Compare the information in document A versus document B on the topic of data security."
- "Summarize our conversation so far and then answer my next question with that context in mind."
- "Search across all knowledge bases and find information about API rate limits."
- "What tools do you have access to and how can you help me with business process automation?"Troubleshooting Ragent
Application fails to start with Redis connection error
Ensure Redis is running and accessible at the host/port configured in application.yml. Start Redis with 'redis-server' and verify connectivity with 'redis-cli ping'. Update spring.redis.host and spring.redis.port in the config to match.
Document ingestion completes but retrieval returns no results
Check that the vector store is properly configured and the embedding model API is reachable. Verify that documents were successfully embedded by checking the vector store table for entries. Ensure the embedding model dimensions match the vector store configuration.
Model circuit breaker opens and all requests fail
The three-state circuit breaker trips when the primary model has repeated failures. Check LLM API credentials and rate limits. Configure a fallback model in the application settings so the circuit breaker can route to a backup model automatically.
Frequently Asked Questions about Ragent
What is Ragent?
Ragent is a Model Context Protocol (MCP) server that 企业级 agentic rag 智能体 - 全链路覆盖文档解析、多路检索、意图识别、问题重写、会话记忆、mcp 工具调用与深度思考。面向真实业务场景，从 0 到 1 完整工程实现。 It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Ragent?
Follow the installation instructions on the Ragent GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Ragent?
Ragent works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Ragent free to use?
Yes, Ragent is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Ragent Alternatives — Similar Coding Agents Servers
Looking for alternatives to Ragent? 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 Ragent 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 Ragent?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.