How does Boston Core ensure data safety?

It guarantees safety by offering read-only access with strictly enforced timeouts and record limits, ensuring secure and controlled interactions with public data for AI systems.

Is Boston Core easy to integrate with AI models?

Yes, it is designed for ease of integration with LLMs and AI agents. It features LLM-friendly outputs, clear contracts, and minimal Python dependencies, making deployment and integration straightforward for AI assistants.

Boston Core

Name: Boston Core
Author: thealphacubicle

0•

Provides large language models and agentic tools with safe, read-only access to Boston's open data portal for exploring trusted civic information.

Boston Core is a collection of Model Context Protocol (MCP) servers developed by the City of Boston's DoIT team. It empowers large language models and autonomous agents with safe, read-only access to Boston's open data portal, enabling staff, residents, and partners to effortlessly explore trusted civic information through conversational interfaces. The project currently features the Boston Open Data server, which wraps the City's CKAN instance, offering a curated toolset for searching datasets, inspecting metadata, and querying DataStore resources with predictable, LLM-friendly outputs.

主な機能

01Curated toolset for dataset discovery, metadata inspection, and DataStore queries

020 GitHub stars

03Read-only access with enforced timeouts and record limits

04Minimal Python dependencies for straightforward deployment

05Exposes Boston's CKAN open data portal

06LLM-friendly outputs for natural-language assistants and autonomous agents

ユースケース

01Integrating Boston's open data into AI assistants and agentic workflows

02Programmatically searching datasets, inspecting metadata, and querying data resources

03Enabling conversational exploration of Boston's civic data