SMKRV MCP Studio
SMKRV MCP Studio is a Docker application for visually constructing MCP (Model Context Protocol) servers from your existing databases. Connect your databases, build SQL-backed tools and resources, design prompt templates, and deploy a fully functional MCP server - all from a visual web interface.Key Features
Connection Manager
PostgreSQL, MySQL, Microsoft SQL Server, ClickHouse, SQLite, Cassandra, Greenplum, Supabase, Snowflake, and BigQuery with encrypted credentials and connection pooling.
Tool Builder
Monaco SQL editor with parameter auto-detection, typed inputs, live preview, and code generation.
Visual Flow Editor
Drag-and-drop canvas for building and managing MCP configurations with template blueprints and batch operations.
One-Click Deploy
Generate and deploy a complete MCP server with real-time log streaming.
Agent MCP Access
45 MCP tools for AI agents (Claude, GPT, Cursor) with temporary tokens and OAuth2 client credentials.
Security
Encryption at rest, TOTP 2FA, Docker network isolation, SSL/TLS with Let’s Encrypt.
Operational Metrics
Per-tool call counts, response times, p95 latency, time-series charts, and queue monitoring.
Change History
Full audit trail for all entity changes with one-click rollback.
Architecture
The shipped product is a single Docker image (smkrv/smkrv-mcp-studio). Inside, supervisord runs five services in one container:
If you build from source, the same services run as a 5-container
docker-compose stack instead (frontend, backend, mcp, agent-mcp, redis) - see Getting Started for the all-in-one Docker Hub image, or the repository’s docker-compose.dev.yml for a dev build from source.
Next Steps
Getting Started
Run SMKRV MCP Studio in minutes with Docker.
User Guide
Learn how to create connections, build tools, and deploy your MCP server.
API Reference
Complete REST API documentation for all endpoints.
Agent MCP
Give AI agents programmatic access to manage your Studio configuration.