Download a shapefile will all the US Census Tracts.

us_census_tracts.py

import geopandas as gpd
import os
import requests
import shutil

# downloading all the ANSI codes for all the states and DC, skipping the territories
states_list = requests.get('https://www2.census.gov/geo/docs/reference/state.txt').text.split('\n')[1:52]

# states_list should =
'''
['01|AL|Alabama|01779775',
 '02|AK|Alaska|01785533',
 '04|AZ|Arizona|01779777',
 '05|AR|Arkansas|00068085',
 '06|CA|California|01779778',
 '08|CO|Colorado|01779779',
 '09|CT|Connecticut|01779780',
 '10|DE|Delaware|01779781',
 '11|DC|District of Columbia|01702382',
 '12|FL|Florida|00294478',
 '13|GA|Georgia|01705317',
 '15|HI|Hawaii|01779782',
 '16|ID|Idaho|01779783',
 '17|IL|Illinois|01779784',
 '18|IN|Indiana|00448508',
 '19|IA|Iowa|01779785',
 '20|KS|Kansas|00481813',
 '21|KY|Kentucky|01779786',
 '22|LA|Louisiana|01629543',
 '23|ME|Maine|01779787',
 '24|MD|Maryland|01714934',
 '25|MA|Massachusetts|00606926',
 '26|MI|Michigan|01779789',
 '27|MN|Minnesota|00662849',
 '28|MS|Mississippi|01779790',
 '29|MO|Missouri|01779791',
 '30|MT|Montana|00767982',
 '31|NE|Nebraska|01779792',
 '32|NV|Nevada|01779793',
 '33|NH|New Hampshire|01779794',
 '34|NJ|New Jersey|01779795',
 '35|NM|New Mexico|00897535',
 '36|NY|New York|01779796',
 '37|NC|North Carolina|01027616',
 '38|ND|North Dakota|01779797',
 '39|OH|Ohio|01085497',
 '40|OK|Oklahoma|01102857',
 '41|OR|Oregon|01155107',
 '42|PA|Pennsylvania|01779798',
 '44|RI|Rhode Island|01219835',
 '45|SC|South Carolina|01779799',
 '46|SD|South Dakota|01785534',
 '47|TN|Tennessee|01325873',
 '48|TX|Texas|01779801',
 '49|UT|Utah|01455989',
 '50|VT|Vermont|01779802',
 '51|VA|Virginia|01779803',
 '53|WA|Washington|01779804',
 '54|WV|West Virginia|01779805',
 '55|WI|Wisconsin|01779806',
 '56|WY|Wyoming|01779807']
 '''

# looping through the states, pulling the shapefiles from the census bureau website.
directory = 'tracks' # define the directory to house the state shapefiles.
os.mkdir(directory)

for state in states_list:
    ids = state.split('|')
    
    url = f'https://www2.census.gov/geo/tiger/GENZ2017/shp/cb_2017_{ids[0]}_tract_500k.zip'

    target_path = f'tracts/{"_".join(ids)}.zip'

    response = requests.get(url, stream=True)
    handle = open(target_path, "wb")
    for chunk in response.iter_content(chunk_size=512):
        if chunk:  # filter out keep-alive new chunks
            handle.write(chunk)
    handle.close()
    
# now that the 'tracts' directory is full of all the shapefiles, we need to 
# consolidate them into one big dataframe.
gdf = gpd.GeoDataFrame()

for filename in os.listdir(directory):
    gdf = gdf.append(gpd.read_file(f'zip://{directory}/{filename}'))
    
    
# putting the GeoDataFame into a zip file
def zip_gdf(gdf, zip_name):    
    os.mkdir(zip_name)
    gdf.to_file(f'{zip_name}/{zip_name}.shp')
    shutil.make_archive(zip_name, 'zip', zip_name)
    shutil.rmtree(zip_name)

zip_gdf(gdf, 'us_census_tracts')