Data AnalysisVerified
Reviewed and published by trentmaziarz, June 30, 2026. Discovered and drafted by our automated research pipeline.
The City of Boston tested an AI agent system called OpenContext, which connects large language models to Boston's public open data portal using the Model Context Protocol (MCP), allowing employees and residents to query city datasets in plain language. As of early 2026, this was in beta and publicly available via Boston's open data showcase.
Details
Boston's Chief Information Officer Santi Garces built OpenContext as a beta MCP server tied to the city's Analyze Boston open data portal. The system receives plain-language questions, automatically identifies the relevant dataset, generates a SQL query, retrieves live data, and returns a formatted answer. Garces launched the MCP server in October 2025 and has made the underlying 'agent skills' open-source, with plans to make the server publicly accessible so any resident can connect their own AI tools to Boston's data.
Products affected
Analyze BostonOpenContext (beta)
Sources & Evidence
News Article
Cite this record
Trace Foundation. (2026). City of Boston: The City of Boston tested an AI agent system called OpenContext, which connects large language models to Boston's public open data portal using the Model Context Protocol (MCP), allowing employees and residents to query city datasets in plain language. As of early 2026, this was in beta and publicly available via Boston's open data showcase (data as of 2026-06-30) [Data set record]. AI Trace. https://www.aitrace.org/r/practice/47a09c3b-6ebb-4375-87a2-20b9c0c753b3. Accessed October 5, 2026.
- Stable link
- https://www.aitrace.org/r/practice/47a09c3b-6ebb-4375-87a2-20b9c0c753b3
- Data as of
- June 30, 2026
- Last verified
- Not recorded
Other practices by City of Boston
Customer SvcThe City of Boston integrated an AI-native platform called Creatio into its 311 non-emergency services system, replacing a legacy system that had operated since 2008. The new platform uses predictive machine learning to recommend next steps in service workflows and generative AI to support natural language interactions between staff and the system.ProductivityThe City of Boston used Google's Gemini 1.5 Pro large language model to automatically generate descriptive titles for more than 1,100 City Council roll call vote records spanning the past 16 years, making historical voting data easier for residents to browse. The city described this as its first public use of generative AI on Boston.gov, launched in May 2024.Customer SvcThe City of Boston deployed a beta AI search feature on Boston.gov, powered by Google Gemini, that gives residents direct answers to questions about city services rather than a list of links. As of December 2025, the feature is live on the 60 most visited pages of Boston.gov.
Have evidence about City of Boston's AI practices? Submit a report.
Report a Sighting →