Mcp Vs Promptql
What is Mcp Vs Promptql
mcp-vs-promptql is a modified MCP server that supports multiple PostgreSQL databases and SQL tools, allowing for enhanced data management and querying capabilities.
Use cases
Use cases include managing control plane data, handling support ticket information, and integrating SQL tools for data analysis and reporting.
How to use
To use mcp-vs-promptql, set up two PostgreSQL databases named ‘control_plane’ and ‘support_tickets’ on localhost:5432, load the provided SQL dumps, customize the server by replacing the ‘index.ts’ file, and configure the server settings before running it.
Key features
Key features include support for multiple databases, enhanced tool descriptions to prevent errors, and the addition of a Python tool for extended functionality.
Where to use
mcp-vs-promptql can be used in data management, application development, and environments requiring complex SQL queries across multiple databases.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Overview
What is Mcp Vs Promptql
mcp-vs-promptql is a modified MCP server that supports multiple PostgreSQL databases and SQL tools, allowing for enhanced data management and querying capabilities.
Use cases
Use cases include managing control plane data, handling support ticket information, and integrating SQL tools for data analysis and reporting.
How to use
To use mcp-vs-promptql, set up two PostgreSQL databases named ‘control_plane’ and ‘support_tickets’ on localhost:5432, load the provided SQL dumps, customize the server by replacing the ‘index.ts’ file, and configure the server settings before running it.
Key features
Key features include support for multiple databases, enhanced tool descriptions to prevent errors, and the addition of a Python tool for extended functionality.
Where to use
mcp-vs-promptql can be used in data management, application development, and environments requiring complex SQL queries across multiple databases.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Content
mcp-vs-promptql
Prepare data
- Go to
datadirectory. - Create two postgres DBs at localhost:5432 -
control_planeandsupport_tickets. - Complete database dumps are given at
control_plane_dump.sqlandsupport_tickets_dump.sql. Load them into the DBs respectively
Customize Postgres MCP Server
Customizations done:
- Support multiple databases and hence multiple SQL tools. By default it’s not possible since postgres MCP always names the tool “query”.
- In the tool description add the DB schema so that Claude doesn’t hallucinate / retry all the time.
- Add python tool (needs
python3in PATH)
Running:
- The
index.tsfile contains the modified server. It won’t compile as-is, you have to: - Clone the original repo: https://github.com/modelcontextprotocol/servers
- Replace the
src/postgres/index.tsfile in there with theindex.tsfile here. - Set the DB URLs correctly in
controlPlaneUrlandsupportTicketsUrlvariables in the file. - Then in that directory run:
npx tsc- Optional: Run
node dist/index.jsto verify the server starts up fine.
Claude Desktop config
Use MCP instructions here: https://modelcontextprotocol.io/quickstart and modify the configuration like below:
Put this in ~/Library/Application Support/Claude/claude_desktop_config.json on Mac:
{
"mcpServers": {
"dbs": {
"command": "node",
"args": [
"<path to cloned servers repo>/src/postgres/dist/index.js"
]
}
}
}Dev Tools Supporting MCP
The following are the main code editors that support the Model Context Protocol. Click the link to visit the official website for more information.
