bonnard-cli
Agent-native analytics. MCP server, dashboards, SDK, and semantic layer CLI.
claude mcp add --transport stdio meal-inc-bonnard-cli npx -y @bonnard/cli mcp
How to use
The Bonnard MCP server is included as part of the Bonnard CLI package. It provides a governed agent-query layer that serves MCP-based requests using the same semantic layer and models you configure with the Bonnard CLI. Once started, you can test the MCP endpoints and validate your deployed agents and prompts against a centralized governance surface. Use the CLI to scaffold and deploy your MCP-enabled setup, then run the MCP server to accept queries and tests via the provided commands.
Typical workflows include initializing a project with bon init, configuring your MCP surface with bon mcp, validating configurations with bon mcp test, and then querying governance-backed outputs through your MCP endpoints. This server is designed to integrate smoothly with the rest of the Bonnard toolkit, so you can rely on a single schema and consistent governance across agents, dashboards, and REST queries.
How to install
Prerequisites:
- Node.js 20 or newer installed on your machine
- npm (comes with Node.js)
Install and run the MCP server:
- Install the Bonnard CLI (or run via npx):
npm i -g @bonnard/cli
or use npx directly without global install
- Start the MCP server (example using npx): npx -y @bonnard/cli mcp
- If you prefer a direct install-and-run approach, you can also initialize a Bonnard project first and then run the MCP server from your project directory: npm i @bonnard/cli npx @bonnard/cli mcp
- Follow the CLI prompts to configure your MCP surface, test it with bon mcp test, and connect agents and dashboards as needed.
Note: The MCP server is accessed via the Bonnard CLI commands, so ensure your environment has network access to any required services (data warehouses, REST APIs, etc.).
Additional notes
Tips and considerations:
- Ensure Node.js 20+ is installed for compatibility with the Bonnard CLI tooling.
- Use bon mcp to set up and test your MCP surface; the test command helps verify governance rules and query routing.
- Keep your credentials and API keys in environment variables or a secrets manager; avoid hard-coding secrets in config files.
- If you encounter issues with npx install-time prompts, you can install the Bonnard CLI globally and run bon mcp locally.
- The MCP surface shares the same metric definitions as other Bonnard components, so changes to the semantic model propagate consistently across agents, dashboards, and REST queries.
Related MCP Servers
mindsdb
Query Engine for AI Analytics: Build self-reasoning agents across all your live data
Wax
Sub-Millisecond RAG on Apple Silicon. No Server. No API. One File. Pure Swift
OpenContext
A personal context store for AI agents and assistants—reuse your existing coding agent CLI (Codex/Claude/OpenCode) with built‑in Skills/tools and a desktop GUI to capture, search, and reuse project knowledge across agents and repos.
mcp-cli
mcpc is a CLI client for MCP. It supports persistent sessions, stdio/HTTP, OAuth 2.1, JSON output for code mode, proxy for AI sandboxes, and much more.
Frappe_Assistant_Core
Infrastructure that connects LLMs to ERPNext. Frappe Assistant Core works with the Model Context Protocol (MCP) to expose ERPNext functionality to any compatible Language Model
shinzo-ts
TypeScript SDK for MCP server observability, built on OpenTelemetry. Gain insight into agent usage patterns, contextualize tool calls, and analyze server performance across platforms. Integrate with any OpenTelemetry ingest service including the Shinzo platform.