MCP Catalogs
Home

bonnard

by bonnard-data·26·Score 44

Bonnard is an agent-native semantic layer providing MCP server capabilities for reliable data querying through AI agents.

databaseai-llmdeveloper-tools
2
Forks
0
Open issues
3 mo ago
Last commit
2d ago
Indexed

Overview

Bonnard provides a self-hosted semantic layer that enables AI agents to query data through the Model Context Protocol. It offers SQL-based metric definitions with caching, access control, and multi-database support. The project includes a Cube semantic layer, pre-aggregation cache, admin UI, deploy API, and health endpoints. It can be deployed as a Docker container and supports various data warehouses including Snowflake, BigQuery, Databricks, PostgreSQL, Redshift, DuckDB, and ClickHouse.

Try asking AI

After installing, here are 3 things you can ask your AI assistant:

you:AI agents querying a consistent semantic layer for analytics and reporting
you:Building data applications with governed metrics and dimensions
you:Embedding structured data capabilities into AI workflows through MCP

When to choose this

Choose Bonnard when you need a self-hosted semantic layer for AI agents with existing database infrastructure and prefer not to use cloud services.

When NOT to choose this

Don't choose Bonnard if you need real-time data updates (it's optimized for analytical queries) or if you prefer a fully managed cloud service over self-hosting.

Tools this server exposes

4 tools extracted from the README
  • query_data

    Query the semantic layer through the MCP server

  • deploy_models

    Deploy updated semantic models to the server

  • check_health

    Check the health status of the Bonnard server

  • view_schema

    Browse and verify the semantic layer schema

Note: Tool names inferred from documentation of MCP server capabilities, CLI commands, and described functionality. No explicit tool names provided in the README.

Comparable tools

cubejs-mcppostgres-mcpsupabase-mcpneon-mcp

Installation

# 1. Scaffold project
npx @bonnard/cli init --self-hosted

# 2. Configure your data source
#    Edit .env with your database credentials

# 3. Start the server
docker compose up -d

# 4. Define your semantic layer
#    Add cube/view YAML files to bonnard/cubes/ and bonnard/views/

# 5. Deploy models to the server
bon deploy

# 6. Connect AI agents
bon mcp

For Claude Desktop / Cursor:

{
  "mcpServers": {
    "bonnard": {
      "url": "https://bonnard.example.com/mcp",
      "headers": {
        "Authorization": "Bearer your-secret-token-here"
      }
    }
  }
}

Compare bonnard with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.