NRTSearch

Provides an MCP interface for Lucene/NRTSearch queries, enabling AI assistants to perform fast, accurate searches.

python

0

GitHub Stars

python

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "tvergilio-nrtsearch-mcp-server": {
      "command": "python",
      "args": [
        "-m",
        "nrtsearch_mcp.server"
      ],
      "env": {
        "LOG_LEVEL": "INFO",
        "NRTSEARCH_MCP_CONFIG": "/path/to/config.json"
      }
    }
  }
}

You expose Lucene/NRTSearch as an MCP server so AI assistants and copilots can query your indexes using natural language or structured Lucene queries. This makes it easy to integrate search into AI workflows, chatbots, and developer tools while keeping full control over configuration and access.

How to use

Start the MCP server locally or in your environment, then connect your MCP client (such as an IDE plugin or AI assistant) to the server to perform searches. You can query any Lucene syntax, request specific fields, enable highlighting, and control how many top results you receive. Use the server to translate natural language questions into Lucene queries, or to run precise searches against your indexes.

How to install

Prerequisites you need on the machine where you run the MCP server:

Python 3.8+ installed on your system.
Git installed for cloning or pulling updates.

Installation steps (one-time or whenever you set up a new environment):

# If you have a repository layout, clone the project
# Replace with the exact source location you use
git clone https://example.com/nrtsearch-mcp-server.git
cd nrtsearch-mcp-server

# Install runtime dependencies (use your preferred method)
pip install -r requirements.txt

# Start the server (choose one of the two options below)

# Option A: Run as a Python module
python -m nrtsearch_mcp.server

# Option B: Use the CLI entrypoint if installed via pip/pipx
nrtsearch-mcp

Configuration and usage patterns

Configure the server via environment variables or a config file. The following settings are commonly used:

LOG_LEVEL to control logging verbosity (default: INFO).
NRTSEARCH_MCP_CONFIG to point to a JSON/YAML config file (optional).
NRTSEARCH_MCP_CONFIG default search might point to a local host and port for simple testing.

Available tools

search

Main tool to perform Lucene queries against the configured NRTSearch index. You provide index, queryText, topHits, optional retrieveFields, and highlight preferences; you receive a list of hits with score, fields, and text content.