meal-inc

meal-inc / bonnard-cli

Public

Agent-native analytics. MCP server, dashboards, SDK, and semantic layer CLI.

www.bonnard.dev agentic-analytics ai-agents analytics bigquery business-intelligence
42
2
100% credibility
Found Feb 18, 2026 at 18 stars 2x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Bonnard CLI lets everyday people define consistent business metrics and categories in simple files, connect them to data storage, and share a unified view for teams, charts, and AI assistants.

How It Works

1
📖 Discover Bonnard

You hear about Bonnard, a simple way to define your key business numbers once and use them anywhere with teams or smart helpers.

2
🚀 Start your project

You set up a new space for your metrics with one easy action, and it even creates guides for your AI coding friends.

3
Connect your data
💼
Use real data

Point to your main data spot like your company's records.

🎮
Try sample data

Jump in with a ready retail example to experiment safely.

4
✏️ Define metrics

You describe what matters like total sales or customer groups in plain files that anyone can understand.

5
✅ Check it works

You test your setup right on your computer to make sure everything adds up perfectly.

6
🌐 Share everywhere

You launch it online so teams, charts, and AI assistants can all use the same trusted numbers.

📊 Insights for all

Now you query smart summaries from anywhere, and your AI helpers understand your business instantly.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 42 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is bonnard-cli?

Bonnard-cli is a TypeScript CLI that deploys an agentic-first, MCP-ready semantic layer in minutes, letting your data flow like water to AI agents, BI tools, and analysts. Define metrics and dimensions once in YAML files, connect warehouses like Snowflake or BigQuery via auto-detection or dbt import, validate locally, and ship a governed API with `bon deploy`. Query from your terminal with JSON or SQL syntax, or hook up Claude and Cursor directly via generated agent configs.

Why is it gaining traction?

Unlike dashboard-first semantic layers retrofitted for AI, Bonnard is agent-native from day one—MCP server ships with native Claude/Cursor/Codex support, auto-generating context files on init. Ultra-fast deploy skips complex setup: detect datasources, test connections, diff changes, and CI/CD with one flag. Metabase analysis tools scan queries for migration insights, making it a practical bridge from BI to agentic workflows.

Who should use this?

Data engineers at dbt-heavy teams needing a quick metrics layer for AI agents. Analytics leads migrating Metabase questions to governed APIs. AI builders prompting Claude/Cursor on production data without warehouse credentials scattered everywhere.

Verdict

Grab it for agentic-first data projects—demo datasource gets you querying instantly, docs and Discord are solid. At 12 stars and 1.0% credibility, it's early but MIT-licensed with TypeScript polish; test before prod. (187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.