OrionBelt Analytics

You run OrionBelt Analytics as an MCP server that analyzes relational schemas and automatically creates ontology mappings while keeping SQL queries safe and well-formed. You can connect to PostgreSQL, Snowflake, or Dremio, generate RDF/OWL ontologies with SQL mappings, and validate or execute queries with built-in safeguards. This enables you to build semantically-aware analytics workflows and produce reliable, fan-trap-free SQL across supported databases.

How to use

You use an MCP client to connect to the OrionBelt Analytics MCP server, choose a workflow, and let the server analyze schemas, generate ontologies, and prepare SQL mappings. Start by connecting to a database, then analyze its schema, generate an ontology with embedded SQL mappings, and optionally run queries, validate them, or generate charts. The server also offers workflow guidance to suggest the next best tool in your analytics sequence, making it easy to follow robust Text-to-SQL pipelines.

How to install

Prerequisites: You need Python 3.13 or higher and a Python environment manager such as uv or a virtual environment.

# Prerequisites
- Python 3.13+
- uv package manager (recommended) or a virtual environment setup

# Step 1: Install dependencies (example flow)
# If using uv
uv sync

# If installing manually
python3.13 -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
pip install -e .

# Step 2: Create configuration
cp .env.template .env  # or create a new .env file with required settings

# Step 3: Start the server (example with uv)
uv run server.py

# Step 4: Verify server is reachable (optional)
curl http://localhost:9000/mcp
"} ]},{

Additional sections

Configuration, security, troubleshooting, and usage notes are provided to help you run OrionBelt Analytics in production or development environments. You can configure database connections, transport mode, output directories, and ontology generation parameters directly in environment variables or through tool parameters. The server includes fan-trap protection, SQL safety checks, and multiple data visualization options through built-in charting tools.

Available tools

connect_database

Connect to PostgreSQL, Snowflake, or Dremio with environment variable fallback. Parameters are optional and can override .env values.

list_schemas

Return an array of available schema names from the connected database.

reset_cache

Clear cached schema analysis and ontology data for the current session.

analyze_schema

Analyze the database schema and return comprehensive table information including relationships.

generate_ontology

Generate RDF/OWL ontology from the database schema with SQL mapping annotations.

suggest_semantic_names

Extract and analyze names from a generated ontology to identify abbreviations and cryptic names for business-friendly improvements.

apply_semantic_names

Apply LLM-suggested semantic names to the ontology, updating labels and adding business descriptions.

load_my_ontology

Load a custom .ttl ontology file from an import folder to use as semantic context.

sample_table_data

Secure data sampling with comprehensive validation.

validate_sql_syntax

Advanced SQL validation with comprehensive analysis.

execute_sql_query

Safe SQL execution with fan-trap protection, pattern analysis, and result validation.

generate_chart

Generate interactive or static charts from SQL results using Plotly and MCP-UI support.

get_server_info

Provide comprehensive server status and configuration information.