Discover
It searches catalogue entries by meaning and keyword, then ranks the strongest sources.
GitHub A context layer for Canadian public data
Explore
open data.
Connect your AI assistant to 75,000 public datasets and statistical tables from Government of Canada, Statistics Canada, Alberta and Ontario sources. Find the right sources, query the data, and build charts, visuals, reports and stories—all traceable to the official source.
LIVE WORKFLOW
0175,000 datasets
0212 candidates
033 datasets
04Chart ready
Local Index75,000 · catalogue metadata
Natural Language12 MCP tools · discovery and analysis
Official Sources4 catalogues · results you can trace
No API keysready to run
The problem
Less hunting through portals.
More answers from data.
A vast amount of public data is available, but getting from a question to an answer often means searching unfamiliar catalogues, decoding spreadsheets and figuring out which resource is current.
opendata.fyi gives your AI assistant the tools to handle that work.
How it works
A traceable path.
Inspect
Before touching the data, it checks fields, sheets, formats and date coverage to choose the right resource.
Query
It filters at the source when possible and streams CSV, Excel, JSON, ZIP, Parquet or PDF only when needed.
Cite
Direct dataset links keep every result traceable and make the original source easy to verify.
What could you ask?
Follow your curiosity.
75,000 datasets and statistical tables
Sample question01 / 06
“How has Alberta’s electricity generation mix changed over time?”
Under the hood
Open-source
MCP server.
opendata.fyi encodes metadata from Canada, Alberta, Ontario and the complete Statistics Canada WDS table inventory into 384-dimensional vectors stored in DuckDB. Each question runs against that shared semantic index while the three CKAN catalogues are searched live. Reciprocal Rank Fusion combines those signals into one ranked result set.
For retrieval, the server chooses the most efficient read-only path. CKAN datastores are filtered server-side, remote files are queried with DuckDB, and the StatCan WDS tool retrieves table metadata, vectors, coordinates and bounded time-series data directly from the official API.
Open source and ready to run
Connect opendata.fyi
to your AI assistant.
Clone the repository, install the Python dependencies, download the latest catalogue index, and add the MCP server to your client.
View setup instructions$ git clone https://github.com/opendatafyi/openmcp.git
$ cd openmcp
$ python3 -m venv venv
$ venv/bin/pip install -r requirements.txt02
Download catalog.duckdb from the latest release and place it in the project root.
03
Connect the server in your MCP client configuration. Full examples are in the repository.
Transparent by design
Optional telemetry.
Off by default.
Catalogue search and embeddings run on your machine. Data requests go directly to the source.
If you opt in, opendata.fyi records a session ID that lasts for one server run, along with server version, tool names, response times, outcomes, general error types, and the public dataset IDs surfaced, inspected or queried. This helps show which tools and datasets get used, where failures happen, and where performance needs attention.
It never records your questions, SQL, filters, complete URLs, detailed errors, file paths or resource contents.
FAQ
Questions,
answered.
An open-source MCP server that helps AI assistants discover and query Canadian public data across federal, Alberta, Ontario and Statistics Canada sources.
opendata.fyi
Find. Build.
Share.
public data with your favorite AI assistant.Federal, Alberta, Ontario and Statistics Canada sources—one traceable workflow.
View source on GitHub