Sphera Credit Data

Canadian rates and market data for commercial real estate lending, as an MCP server for AI assistants and as open JSON and CSV files.

Connect an AI assistant

Add this URL as a remote MCP server:

https://data.spheracredit.com/mcp

Browsing the catalogue and latest values works right away. The first time your assistant pulls a rate history, it asks you to sign in with Google, Microsoft or a code sent to your email. Access is free.

Claude
  1. In Claude, open Customize, then Connectors.
  2. Add a custom connector and paste the URL above.
  3. Ask for a rate history. The first time, Claude shows a Connect button: sign in with Google, Microsoft or a code sent to your email.

On Team and Enterprise plans, an Owner adds the connector once, then each member connects.

Claude Code

Add the server, then sign in from your terminal.

claude mcp add --transport http sphera-data https://data.spheracredit.com/mcp
claude mcp login sphera-data
ChatGPT
  1. Open Settings, select Security and login, and turn on Developer mode.
  2. Go to Plugins, select the plus button, name it Sphera Credit Data and paste the URL above.
  3. Ask for a rate history and sign in with Google, Microsoft or a code sent to your email when ChatGPT asks.

Developer mode is on the web, and its availability depends on your plan and workspace policy.

Codex

Add the server. Codex opens the sign-in page right away.

codex mcp add sphera-data --url https://data.spheracredit.com/mcp

What you can ask

Tools

Your assistant chooses between two tools. You do not call them yourself.

List Canadian rate and market series

No sign-in needed
list_series

Lists the available series (Government of Canada bond yields, CORRA, Canada Mortgage Bond yields and spreads, policy and prime rates, bank non-residential mortgage rates, inflation, USD/CAD) with each series' id, unit, frequency, source, first date and latest value. Use it to find ids for get_series and to answer questions about the latest value. To explain how a series is defined, sourced or calculated, call it with include_methodology set to true and a query or category that selects that series; each series then also carries its methodology (source, definition, calculation, timing, caveats and references).

querystring, optional
Words to match against series ids, names and descriptions, e.g. "cmb 5-year" or "corra".
categorystring, optional
Restrict results to one category. Values: bonds, corra, cmb, policy, mortgage, inflation, fx.
include_methodologyboolean, optional
Add each series' methodology. Narrow the results with query or category first.

Get series history

Sign-in required
get_series

Returns date-aligned observations for one to ten series ids, optionally within a date range and converted to weekly, monthly or annual values. Each series comes with its unit, source citation, whether it is a derived estimate, and its methodology: source, definition, calculation, timing, caveats and references. Large requests are converted to a coarser frequency to fit, and the result says so and links the full CSV.

seriesarray, required
One to ten series ids from list_series, e.g. ["goc-5y", "cmb-5y-estimate"].
start_datestring, optional
First date to include, YYYY-MM-DD.
end_datestring, optional
Last date to include, YYYY-MM-DD.
frequencystring, optional
Convert to this frequency. Default native. Mixed frequencies are always converted to the coarsest one. Values: native, weekly, monthly, annual.
aggregationstring, optional
How to convert to a coarser frequency: last value in the period (default) or the period mean. Values: last, mean.

Limits

Open data

Catalogue: /api/v1/catalog.json. Each series is also available as /api/v1/series/{id}.csv, which Excel can refresh through Data, From Web.

Source: Bank of Canada. Series marked derived are Sphera estimates computed from Bank of Canada data. Privacy notice Support Terms of use