DataBridge AI - Community Edition

Open-source, MCP-native data reconciliation engine for comparing, profiling, and managing data quality.

Overview

DataBridge AI Community Edition is a free, open-source data reconciliation toolkit built on the Model Context Protocol (MCP). It provides essential tools for:

• Data Comparison - Hash-based row comparison, orphan detection, conflict identification
• Fuzzy Matching - Find approximate matches between datasets using RapidFuzz
• Data Profiling - Statistical analysis and quality metrics for your data
• PDF/OCR Extraction - Extract text from PDFs and images
• dbt Integration - Generate dbt projects from your data
• Data Quality - Create and run data validation rules

Installation

# Basic installation
pip install databridge-ai

# With PDF support
pip install databridge-ai[pdf]

# With OCR support
pip install databridge-ai[ocr]

# With all optional dependencies
pip install databridge-ai[all]

Quick Start

As an MCP Server

DataBridge AI works as an MCP server, making its tools available to AI assistants like Claude:

# Run the MCP server
databridge-mcp

Using the Dashboard

# Start the web dashboard
databridge-ui

Then open http://localhost:5050 in your browser.

The dashboard includes a dedicated Hierarchy Builder page for visually creating and managing multi-level hierarchy projects, importing from CSV, and exporting deployment scripts.

Python API

from databridge_ai import load_csv, profile_data, fuzzy_match_columns

# Load and profile a CSV file
result = load_csv("data.csv")
profile = profile_data("data.csv")

# Find fuzzy matches between two files
matches = fuzzy_match_columns(
    source_file="source.csv",
    target_file="target.csv",
    source_column="name",
    target_column="customer_name",
    threshold=80
)

Available Tools (~106 tools)

Editions

DataBridge AI is available in four editions:

Upgrade to Pro

# Set your license key
export DATABRIDGE_LICENSE_KEY="DB-PRO-YOURKEY-20260101-signature"

# Install Pro (from GitHub Packages)
pip install databridge-ai-pro --extra-index-url https://ghp_TOKEN@raw.githubusercontent.com/datanexum/DATABRIDGE_AI/main/

# Install Pro Examples (tests & tutorials, requires Pro key)
pip install databridge-ai-examples                # CE tests + beginner tutorials
pip install databridge-ai-examples[pro]           # + Pro tests + advanced tutorials

License Key Format: DB-{TIER}-{CUSTOMER_ID}-{EXPIRY}-{SIGNATURE}

Contact sales@databridge.ai for pricing and license keys.

Configuration

Create a .env file in your project root:

# Database connection (optional)
DATABRIDGE_DATABASE_URL=postgresql://user:pass@localhost/db

# OCR settings (optional)
DATABRIDGE_TESSERACT_PATH=/usr/bin/tesseract

# Fuzzy matching threshold (default: 80)
DATABRIDGE_FUZZY_THRESHOLD=80

# Max rows to display (default: 10)
DATABRIDGE_MAX_ROWS_DISPLAY=10

Plugin System

Extend DataBridge AI with custom plugins:

plugins/
├── my_plugin/
│   ├── __init__.py
│   └── mcp_tools.py  # Must have register_tools(mcp)

# plugins/my_plugin/mcp_tools.py
def register_tools(mcp):
    @mcp.tool()
    def my_custom_tool(param: str) -> str:
        """My custom tool description."""
        return f"Processed: {param}"

Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

License

MIT License - see LICENSE for details.

Links

• Documentation: github.com/datanexum/DATABRIDGE_AI/wiki
• Commercialization Guide: docs/COMMERCIALIZATION.md
• Issues: github.com/datanexum/DATABRIDGE_AI/issues
• Pro Features: Pro Features Wiki