# mcp

An MCP server that translates natural language prompts into MongoDB aggregations and MySQL queries directly inside your agent workflow.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 6
- Author: florentine-ai
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/florentine-ai/mcp
- Updated: 2026-03-09
- HTML: https://affpapa.org/en/nevibecode/mcp/florentine-ai-mcp

## What it does

This server connects coding assistants to MongoDB and MySQL instances via Florentine.ai. When you ask questions in conversational English, it maps intent into database queries and returns records or calculations back to your chat. The package exposes two core MCP tools: `florentine_list_collections` to inspect accessible collections with field types, and `florentine_ask` to execute queries.

The architecture includes multi-tenant data separation and key exclusion to hide sensitive fields. Setup requires Node.js 18.0.0 or newer, a `FLORENTINE_TOKEN` credential, and an LLM API key from OpenAI, Google, Anthropic, or Deepseek. Configuration operates via `--mode static` for standard desktop clients or `--mode dynamic` for custom integrations. Without at least one activated collection in your Florentine dashboard, queries will fail immediately.

## Requirements
- Needs: Node.js (commands run via npx)
- Environment variables: `FLORENTINE_TOKEN`, `LLM_SERVICE`, `LLM_KEY`, `SESSION_ID`, `RETURN_TYPES`, `REQUIRED_INPUTS` — keys and tokens: ask the human for the values, never commit them or put them in shared files.

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add --env FLORENTINE_TOKEN=<FLORENTINE_TOKEN> --env LLM_SERVICE=<LLM_SERVICE> --env LLM_KEY=<LLM_KEY> --env SESSION_ID=<SESSION_ID> --env RETURN_TYPES=<RETURN_TYPES> --transport stdio florentine-ai-mcp -- npx -y @florentine-ai/mcp
```
Replace <…> with your own values. 1 more environment variable is in the server README

### Codex

**Platform CLI**
```bash
codex mcp add florentine-ai-mcp --env FLORENTINE_TOKEN=<FLORENTINE_TOKEN> --env LLM_SERVICE=<LLM_SERVICE> --env LLM_KEY=<LLM_KEY> --env SESSION_ID=<SESSION_ID> --env RETURN_TYPES=<RETURN_TYPES> -- npx -y @florentine-ai/mcp
```
Replace <…> with your own values. 1 more environment variable is in the server README

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "florentine-ai-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "@florentine-ai/mcp"
            ],
            "env": {
                "FLORENTINE_TOKEN": "<FLORENTINE_TOKEN>",
                "LLM_SERVICE": "<LLM_SERVICE>",
                "LLM_KEY": "<LLM_KEY>",
                "SESSION_ID": "<SESSION_ID>",
                "RETURN_TYPES": "<RETURN_TYPES>"
            }
        }
    }
}
```
Replace <…> with your own values. 1 more environment variable is in the server README

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "florentine-ai-mcp": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "@florentine-ai/mcp"
            ],
            "env": {
                "FLORENTINE_TOKEN": "<FLORENTINE_TOKEN>",
                "LLM_SERVICE": "<LLM_SERVICE>",
                "LLM_KEY": "<LLM_KEY>",
                "SESSION_ID": "<SESSION_ID>",
                "RETURN_TYPES": "<RETURN_TYPES>"
            }
        }
    }
}
```
Replace <…> with your own values. 1 more environment variable is in the server README

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add -e FLORENTINE_TOKEN=<FLORENTINE_TOKEN> -e LLM_SERVICE=<LLM_SERVICE> -e LLM_KEY=<LLM_KEY> -e SESSION_ID=<SESSION_ID> -e RETURN_TYPES=<RETURN_TYPES> florentine-ai-mcp npx -y @florentine-ai/mcp
```
Replace <…> with your own values. 1 more environment variable is in the server README

## When to use it
- Query MongoDB and MySQL collections using conversational prompts
- Inspect active database schemas and column types via florentine_list_collections
- Enforce tenant isolation across multi-tenant database environments
- Mask sensitive keys from database query responses

## Example prompts
- Run `florentine_list_collections` to inspect active database collections and their schema types
- Use `florentine_ask` to calculate total revenue per customer category in MongoDB
- Ask `florentine_ask` to fetch recent signups with pending verification from MySQL

## Similar
- [World Monitor](https://affpapa.org/en/nevibecode/mcp/world-monitor.md) — MCP server · ★ 89609
- [Scrapling MCP Server](https://affpapa.org/en/nevibecode/mcp/scrapling-mcp-server.md) — MCP server · ★ 89899
- [MCP Toolbox for Databases](https://affpapa.org/en/nevibecode/mcp/googleapis-mcp-toolbox-for-databases.md) — MCP server · ★ 16730
- [PostHog MCP Server](https://affpapa.org/en/nevibecode/mcp/posthog-mcp-server.md) — MCP server · ★ 40459
- [dbx](https://affpapa.org/en/nevibecode/mcp/dbx.md) — MCP server · ★ 27844
- [Metabase](https://affpapa.org/en/nevibecode/mcp/metabase.md) — MCP server · ★ 49863

## Other versions with the same name
- [mcp · kubeshark](https://affpapa.org/en/nevibecode/mcp/kubeshark-mcp.md) — ★ 12110
- [mcp · butterbase-ai](https://affpapa.org/en/nevibecode/mcp/butterbase-ai-mcp.md) — ★ 3031
- [mcp · medplum](https://affpapa.org/en/nevibecode/mcp/mcp.md) — ★ 2766
- [mcp · sceneview](https://affpapa.org/en/nevibecode/mcp/sceneview-mcp.md) — ★ 1349
- [mcp · buoy-gg](https://affpapa.org/en/nevibecode/mcp/buoy-gg-mcp.md) — ★ 691

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md