Skip to main content
Version: v5

Quick start — User

This walkthrough uses FRED, the Federal Reserve Bank of St. Louis economic database. In V5 each provider package adds its own namespace to obb, so FRED commands live under obb.fred.

1. Install​

pip install openbb-fred

openbb-fred depends on openbb-core with its pandas extra, and openbb-core provides the openbb package that from openbb import obb imports. Other ways to install are covered in Installation.

The first import builds the Python Interface for the installed extensions, which takes a moment and prints Building.... In containers and CI, run openbb-build right after installing instead; see Packaging and static assets.

2. Set the FRED API key​

The provider declares one credential, fred_api_key. Request a key from your account at fred.stlouisfed.org under "My Account", then "API Keys".

Export it in the shell that runs Python:

export FRED_API_KEY="YOUR_FRED_KEY"

Or store it in ~/.openbb_platform/user_settings.json, which is created on the first import:

{
  "credentials": {
    "fred_api_key": "YOUR_FRED_KEY"
  },
  "preferences": {},
  "defaults": {
    "commands": {}
  }
}

If both are set, the JSON value is used. Configure credentials explains the other options and how the REST API and MCP server read keys.

3. Make a call​

from openbb import obb

print(obb.fred.economy)

Printing a namespace lists its commands and sub-namespaces, including fred_series. Request a series by its FRED ID:

gdp = obb.fred.economy.fred_series(symbol="GDP", start_date="2015-01-01")

print(gdp.provider)
print(gdp.extra["results_metadata"]["GDP"]["title"])
print(gdp.to_dataframe())

The call returns an OBBject. results holds one record per date, extra["results_metadata"] carries each series' title, units, and frequency, and to_dataframe() returns a pandas DataFrame indexed by date. See The OBBject envelope for the other fields and methods, and the reference page for every parameter.

4. Pick an output type​

For a one-off conversion, call a method such as to_dataframe(), to_dict(), or to_llm() on the result. To change what every command returns in this session, set the preference:

obb.user.preferences.output_type = "dataframe"

Accepted values are "OBBject" (the default), "dataframe", "numpy", "dict", "llm", and "polars". Output types lists what each returns and which packages it needs.

5. Verify​

In a new session, confirm the key is loaded and the call reaches FRED. The key prints as asterisks, and the provider is fred.

from openbb import obb

print(obb.user.credentials.fred_api_key)
print(obb.fred.economy.fred_series(symbol="GDP", limit=5).provider)

A Missing credential 'fred_api_key' error means the key was not found; check the variable name or the JSON file. From here, the User Guide covers defaults, parameters, warnings, and errors, and the REST API quickstart serves the same commands over HTTP.

On this page