bonnard-cli

Overview

Bonnard is an open-source agentic schema CLI optimized for Claude, Gemini, Codex, and Copilot. It serves as a semantic layer that connects data warehouses to various surfaces including AI agents through MCP, dashboards in markdown format, and internal tools via REST API. The platform ensures consistent metrics definitions across all consumers, addressing the common issue of metric drift between different tools. It supports major data warehouses like Snowflake, BigQuery, Databricks, and PostgreSQL, with auto-detection capabilities for existing schemas and relationships.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• bon_init

  Scaffold a new project with agent configs

  Try:Initialize a new Bonnard projectTry:Set up a new semantic layer project
• bon_datasource_add

  Connect a data source or add demo data

  Try:Connect to Snowflake data warehouseTry:Add the retail demo dataset
• bon_validate

  Validate YAML syntax locally

  Try:Validate my model definitionsTry:Check for YAML syntax errors
• bon_deploy

  Deploy to production with a version message

  Try:Deploy with message 'Q2 metrics update'Try:Push my models to production
• bon_pull

  Download deployed models to local project

  Try:Pull latest production modelsTry:Update local models with deployed versions
• bon_query

  Run queries (JSON or SQL) against the semantic layer

  Try:Query total revenue by product categoryTry:Get customer count from orders
• bon_schema

  Explore deployed measures, dimensions, and views

  Try:Show all available metricsTry:List dimensions for the orders view
• bon_dashboard_dev

  Preview a markdown dashboard locally

  Try:Preview my sales dashboardTry:Test the dashboard before deployment
• bon_dashboard_deploy

  Deploy a markdown dashboard

  Try:Deploy my sales dashboardTry:Publish dashboard to production
• bon_mcp

  MCP server setup instructions

  Try:Set up MCP serverTry:Configure Bonnard for agent integration
• bon_keys_create

  Create a publishable or secret API key

  Try:Create an API key for my applicationTry:Generate secret key for SDK access
• bon_docs

  Browse documentation from the CLI

  Try:Show documentation for modelingTry:Get CLI reference guide

Comparable tools

Installation

npm install -g @bonnard/cli

bon mcp

{
  "mcpServers": {
    "bonnard": {
      "command": "npx",
      "args": ["@bonnard/cli", "mcp"]
    }
  }
}

FAQ

What data sources does Bonnard support?

Bonnard supports major data warehouses including Snowflake, Google BigQuery, Databricks, PostgreSQL, Amazon Redshift, and DuckDB. It also integrates with data tools like dbt, Dagster, Prefect, Airflow, and various BI tools.

How does Bonnard ensure metric consistency?

Bonnard provides a unified semantic layer where all consumers (AI agents, dashboards, SDK) query the same governed metrics defined in one schema, eliminating metric drift across different tools.

Compare bonnard-cli with