MCP Server Builder
Enables users to generate, deploy, and register a read-only MCP server from database connection parameters without writing code.
README
🛠️ MCP Server Builder
Generate, deploy, and register a read-only MCP server from user-provided
database parameters — no code required. A non-engineer connects a new data
source end to end in minutes; every write (INSERT/UPDATE/DELETE) is refused.
The six-step flow
- Collect connection parameters (user ID, password, server/IP, port) — password masked.
- Test the connection (
SELECT 1) with a live status indicator; clear re-prompt on failure. - Generate a SELECT-only MCP server (sqlglot-validated).
- Deploy it (runs over HTTP).
- Register it with the host application (writes
generated_servers/registry.json). - Query your data through the new server — writes are refused.
Architecture
frontend/app.py— Streamlit wizard UI (talks to the backend over HTTP).backend/api.py— FastAPI service (test / deploy / register / query).backend/{connection,generator,deploy}.py+templates/mcp_template.py— build logic.
Run locally
Two processes (recommended for development):
python -m uvicorn backend.api:app --port 8000 --reload
streamlit run frontend/app.py
Or a single process — the frontend auto-starts the backend in a background thread if nothing is already listening on port 8000:
streamlit run frontend/app.py
Deploy free on Streamlit Community Cloud
The app is single-container ready: frontend/app.py launches the FastAPI
backend in-process (a daemon thread, started once via @st.cache_resource),
so the whole six-step flow runs inside the one process Streamlit Cloud gives you.
- Push this repo to public GitHub.
- Go to https://share.streamlit.io → New app → select the repo.
- Set Main file path to
frontend/app.py. - Deploy. You get a public
*.streamlit.appURL.
requirements.txt (Python deps) and packages.txt (system unixodbc) are
installed automatically.
Hosting notes / limits (free tier)
- Only publicly-reachable databases work. The container can't reach a DB on
your laptop (
127.0.0.1). Use a cloud DB — e.g. free-tier Neon/Supabase (Postgres) or TiDB Cloud/PlanetScale (MySQL). - SQL Server needs Microsoft's ODBC driver, which Streamlit Cloud can't install — use MySQL/PostgreSQL/TiDB for the hosted build (or self-host for MSSQL).
- The app sleeps when idle and storage is ephemeral: generated servers
and
registry.jsondon't persist across restarts (they regenerate on next use). - The registered MCP endpoints are container-internal (
localhost) — reachable by this platform (which is the host application for querying), not by an external Claude Desktop. That matches the scenario: teams connect and query through the platform.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.