The Florentine.ai MCP Server enables natural language querying for MongoDB and MySQL data, allowing integration into custom AI agents or desktop applications.
From the registry: MCP server for Florentine.ai - Natural language to MongoDB aggregations

Please install the `mcp-florentineai` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Florentine.ai account and API key** — Create a free account at app.florentine.ai/signup, connect a database, activate at least one collection/table, and copy the API key from the dashboard. (https://app.florentine.ai/dashboard)
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["-y","@florentine-ai/mcp","--mode","static"]
- required environment variables:
- `FLORENTINE_TOKEN`: Your Florentine.ai API key (example: `<your-api-key>`)
- optional environment variables:
- `LLM_SERVICE`: LLM provider: openai, google, anthropic, or deepseek (only needed if not stored in Florentine.ai account) (example: `openai`)
- `LLM_KEY`: API key for your LLM provider (only needed if not stored in Florentine.ai account) (example: `<your-llm-api-key>`)
- `SESSION_ID`: Session ID for server-side chat history (example: `<uuid>`)
- `RETURN_TYPES`: Stringified JSON array of return types (query, result, answer) (example: `["result"]`)
- `REQUIRED_INPUTS`: Stringified JSON for multi-tenant data filtering
Note: Natural language querying for MongoDB and MySQL via Florentine.ai. Bring your own LLM key (OpenAI/Google/Anthropic/Deepseek). Use --mode static for Claude Desktop.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.FLORENTINE_TOKENrequiredYour Florentine.ai API key.MCP server for interacting with the Supabase platform