QueryNest

A powerful MongoDB query and analysis MCP server

python

1

GitHub Stars

python

Language

6 months ago

First Indexed

2 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": {
    "niuzaishu-querynest": {
      "command": "uvx",
      "args": [
        "--from",
        ".",
        "--no-cache",
        "querynest-mcp"
      ],
      "env": {
        "QUERYNEST_LOG_LEVEL": "INFO",
        "QUERYNEST_CONFIG_PATH": "/path/to/config.yaml"
      }
    }
  }
}

QueryNest is a Python-based MCP server that enables intelligent querying across multiple MongoDB instances. You can describe what you want in natural language, and the system translates it into MongoDB-native queries, optimized for performance, with safeguards like read-only access and query limits. It automatically discovers MongoDB instances, analyzes schemas, and provides interactive query generation and confirmation through MCP-compliant tooling.

How to use

You connect to QueryNest through an MCP client and perform a sequence of operations to locate databases, analyze collections, and generate queries from natural language descriptions. Start by discovering available MongoDB instances, then inspect databases and collections to understand their structure. When you have a natural language query, generate a corresponding MongoDB query, review it, and confirm execution to obtain results. The system handles authentication, applies safety constraints, and can suggest indexing or optimization opportunities to improve performance.

How to install

# Prerequisites
# - Python 3.8+
# - MongoDB 4.0+
# - Optional: Redis for caching

# Step 1: Install the uv tool
pip install uv

# Step 2: Install and run from project directory (recommended)
cd /path/to/QueryNest
uvx --from . --no-cache querynest-mcp

# Step 3: (Alternative) set a config path and start
export QUERYNEST_CONFIG_PATH=/path/to/config.yaml
uvx --from /path/to/QueryNest --no-cache querynest-mcp

# Step 4: Verify startup
uvx --from . --no-cache querynest-mcp --help

Additional sections

Configuration, security, and usage notes are provided below. You will configure multiple MongoDB instances, ensure read-only access, and enable data masking and auditing to protect sensitive information. Use the MCP tooling to discover instances and databases, analyze collections, and generate queries from natural language descriptions. Environment variables control configuration paths and logging, and the system exposes health and performance metrics for monitoring.

Available tools

discover_instances

Finds and lists all available MongoDB instances, including health and basic stats.

discover_databases

Lists all databases within a specified MongoDB instance, with options to include collections and exclude system databases.

analyze_collection

Analyzes the structure and field information of a specified collection, with semantic and example data if requested.

manage_semantics

Manages business semantic information for fields and collections, enabling semantic consistency across queries.

generate_query

Generates a MongoDB query from a natural language description, tailored to a specific instance/database/collection.

confirm_query

Executes the generated query, returns results, and can provide explain plans for optimization.