Chicago CPS Data

Provides access to local CPS data and embeddings via two query tools for school-neighborhood data and school website context.

typescript

0

GitHub Stars

typescript

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "mdagost-mcp-cps-data": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/mcp-cps-data",
        "run",
        "mcp-cps-data",
        "--sqlite-path",
        "/path/to/cps_crawler.db",
        "--lancedb-path",
        "/path/to/embeddings.lancedb"
      ]
    }
  }
}

This MCP server exposes a local SQLite database and a LanceDB vector store containing Chicago Public Schools information, making it easy to run queries against local data and retrieve contextual results for building CPS-focused applications.

How to use

You run this MCP server as a local process and connect to it with an MCP client to invoke the available tools. Two tools are exposed: a query tool that retrieves school data by neighborhood and a website context tool that fetches context about CPS websites. Use the client to pass the required parameters to each tool and obtain structured results that you can display or analyze in your application.

How to install

Prerequisites you need before starting are a runtime to execute MCP servers and a local environment with the data files ready.

Prepare your environment and data paths. Ensure you have access to your local CPS data files and embeddings. You will reference these paths when starting the server.
Create a directory for the MCP server work. This directory will be used by the server runtime to load your data.
Start the MCP server using the runtime with the appropriate arguments. The command below runs the server in stdio mode and points to your local data files. Replace the paths with your actual locations.

uv \
  --directory "/path/to/mcp-cps-data" \
  "run" 
  "mcp-cps-data" \
  --sqlite-path "/path/to/cps_crawler.db" \
  --lancedb-path "/path/to/embeddings.lancedb"

Additional setup notes

If you are debugging, you can use the MCP Inspector tool to facilitate troubleshooting and to view a debugging URL in your browser. This is especially helpful when you are validating tool parameters and data responses during development.

Configuration and runtime details

Configuration is provided locally via a development/unpublished servers configuration. The MCP server is identified as mcp-cps-data and is started with the runtime that loads the local SQLite database and LanceDB vector store.

Available tools

query_schools_and_neighborhoods

Executes a SELECT query on the schooltoneighborhood table to fetch school names and neighborhoods; requires a parameter query.

query_school_websites

Queries CPS school websites for contextual information relevant to a given question; requires a parameter question and may include an optional school_name.