NRT Search

NRT Search is an MCP server that provides near real-time search, document indexing, and querying capabilities implemented in Java. Built using Spring and distributed as a Java archive, it exposes search engine mechanisms directly to Model Context Protocol clients via standard input/output transport. Developers, search engineers, and automated AI agents use it to bridge conversational assistants with local or remote search backends. By integrating this server into an MCP environment, users enable AI models to index document collections and run low-latency retrieval operations within active chat workflows. The server executes directly on the Java runtime using Spring profiles to manage standard I/O communication, removing the need for external web service layers during local LLM interactions. It offers an effective approach for integrating document search workflows into developer environments, allowing large language models to inspect and query structured text repositories accurately.

Category: Files, Documents & PDFs

Tags: document-retrieval, indexing, java, search

Visit NRT Search

How to install and configure NRT Search

1. Build the application or obtain the packaged JAR file located at ./build/libs/nrtsearch-mcp-0.0.1.jar from the project repository at https://github.com/waziqi89/nrtsearch-mcp. 2. Open your Claude Desktop configuration file (claude_desktop_config.json). 3. Add the server under the mcpServers section: json { "mcpServers": { "nrtsearch": { "command": "java", "args": [ "-jar", "-Dspring.profiles.active=stdio", "./build/libs/nrtsearch-mcp-0.0.1.jar" ] } } } 4. Save the configuration and restart Claude Desktop. 5. Optionally inspect the server using the MCP Inspector: npx @modelcontextprotocol/inspector java -Dspring.profiles.active=stdio -jar ./build/libs/nrtsearch-mcp-0.0.1.jar

What you can do with NRT Search

• Indexing text documents and dynamic data sources directly from an AI assistant interface for rapid access. - Querying local search repositories with low latency to retrieve contextually relevant passages during analysis. - Testing and inspecting Java-based MCP search tooling using the standard Model Context Protocol inspector CLI. - Automating near real-time document search retrieval pipelines within local Claude Desktop developer environments.

Key facts

• https://github.com/waziqi89/nrtsearch-mcp
• Files, Documents & PDFs, Web Search & Research
• document-retrieval, indexing, java, search

Part of MCP Servers

What is NRT Search?

NRT Search is a Model Context Protocol server implemented in Java. It allows AI clients like Claude Desktop to connect to a near real-time search engine backend for indexing and querying documents over standard input and output.

How do I install NRT Search?

To run NRT Search, ensure you have Java installed. Build or locate the application JAR at ./build/libs/nrtsearch-mcp-0.0.1.jar, then add it to your Claude Desktop configuration file specifying the java command with the stdio Spring profile.

Which MCP clients work with NRT Search?

NRT Search works with Claude Desktop and any client that supports standard I/O (stdio) transport for MCP servers. You can also test and inspect its functions using the official MCP Inspector tool via npx.

What can NRT Search do?

NRT Search enables automated assistants to manage document indexing and perform near real-time search queries. It acts as an integration bridge between Java-based search systems and AI conversational tools.