An unofficial screen-scraping API for Mint.com.
Please join us on Discord to get help or just chat with fellow mintapi users :)
Ensure you have Python 3 and pip (easy_install pip
) and then:
pip install mintapi
mintapi
scrapes Mint.com by navigating a Chrome browser (or Chromium) just as a human would. Once logged in, the API allows programatic access to various Mint REST APIs. Selenium/WebDriver is used to accomplish this, and specifically, ChromeDriver under the hood. mintapi
will download the latest stable release of chromedriver, unless --use_chromedriver_on_path is given. NOTE: You must have Chrome or Chromium installed, on the stable
track, and be up-to-date! If you run into a SessionNotCreatedException
about "ChromeDriver only supports Chrome version XX", you need to update Chrome.
From the command line, the most automated invocation will be:
mintapi --keyring --headless [email protected]
This will store your credentials securely in your system keyring, and use a
headless (invisible) browser to log in and grab the account data. If this triggers
an MFA prompt, you'll be prompted on the command line for your code, which by default
goes to SMS unless you specify --mfa-method=email
. This will also persist a browser
session in $HOME/.mintapi/session to avoid an MFA in the future, unless you specify --session-path=None
.
If you wish to simplify the number of arguments passed in the command line, you can use a configuration file by specifying --config-file
. For arguments such as --transactions
, you can add a line in your config file that says transactions
. For other arguments that have input, such as --start-date
, you would add a line such as start-date=10/01/21
. There are two exceptions to what you can add to the config file: email and password. Since these arguments do not include --
, you cannot add them to the config file.
If you're running mintapi in a server environment on an automatic schedule, consider running mintapi in headless mode if you don't need to see the login workflow. In addition, you'll want to use your distribution's package manager to install chromium and chromedriver. Make sure your distribution is up-to-date and then install/update Chromium (debian-family example): apt install chromium-browser chromium-chromedriver
. Then use the option use_chromedriver_on_path
either through the CLI or the python api so that mintapi doesn't try to find a matching chromedriver.
If you need to download the chromedriver manually, be sure to get the version that matches your chrome version and make the chromedriver available to your python interpreter either by putting the chromedriver in your python working directory or inside your PATH
as described in the python selenium documentation.
When running this inside of a cron job or other long-term automation scripts, it might be helpful to specify chrome and chromedriver executables so as not to conflict with other chrome versions you may have. Selenium by default just gets these from your PATH
environment variable, so customizing your environment can force a deterministic behavior from mintapi. To use a different browser besides Chrome or Chromium, see the python api. Below are two examples.
If I wanted to make sure that mintapi used the chromium executable in my /usr/bin directory when executing a cron job, I could write the following cron line:
0 7 * * FRI PATH=/usr/bin:$PATH mintapi --headless [email protected] my_password
where prepending the /usr/bin path to path will make those binaries found first. This will only affect the cron job and will not change the environment for any other process.
You can also use the docker image to help manage your environment so you don't have to worry about chrome or chromedriver versions. There are a few caveats:
- Headless mode is recommended. GUI works but introduces the need to configure an X11 server which varies with setup. Google is your friend.
- Almost always use the flag
--use-chromedriver-on-path
as the chrome and chromedriver built into the docker image already match and getting the latest will break the image. - If you want to persist credentials or your chrome session, you'll need to do some volume mounting.
To use the image:
docker run --rm --shm-size=2g ghcr.io/mintapi/mintapi mintapi [email protected] my_password --headless --use-chromedriver-on-path
You can do a similar thing in windows by executing the following in Powershell.
$ENV:PATH = "C:\Program Files\Google\Chrome;$ENV:PATH"
mintapi --headless john@example.com my_password
As of v2.0, mfa_method
is only required if your login flow presents you with the option to select which Multifactor Authentication Method you wish to use, typically as a result of your account configured to accept different methods.
If mintapi
detects that your Mint account uses IMAP and your email host provides IMAP access, you can specify your IMAP login details. This will automate the retrieval of the MFA code from your email and entering it into Mint. If you use IMAP in conjunction with keyring
, then you can store your IMAP password (imap-password
) in keyring. To do so, simply omit imap-password
and you will initially be prompted for the password associated with your IMAP account. Then, on subsequent uses of your IMAP account, you will not have to specify your password.
If mfa-method
is soft-token then you must also pass your mfa-token
. The mfa-token
can be obtained by going to your mint.com settings and clicking on 'Intuit Account'. From there go to Sign In & Security -> Two-step verification. From there, enable the top option however you wish (either text or email is fine). After that, start the process to enable the Authenticator app option and when you get the part where you see the QR code, copy the manual setup code that appears next to it. Careful where you store this as it allows anyone to generate TOTP codes. This is the token that you will pass to mfa-token
in either the python api or from the command line.
While Mint supports authentication via Voice, mintapi
does not currently support this option. Compatability with this method will be added in a later version.
As of v2.0, MintAPI supports returning multiple types of data in one call to MintAPI. When exporting multiple data types, you can either send it directly to stdout
or you can export to a data file. What MintAPI will do with your specified filename is add a suffix based on the type of data you are exporting. The following table outlines the option selected and its corresponding suffix:
Option | Suffix |
---|---|
accounts | account |
budgets | budget |
transactions | transaction |
categories | category |
investments | investment |
net-worth | net_worth |
credit-score | credit_score |
credit-report | credit_report |
For example, if you specify current
as your filename, format as csv, and you export account
and transaction
, then you will receive two files: current_account.csv
and current_transaction.csv
.
From python, instantiate the Mint class (from the mintapi package) and you can
make calls to retrieve account/budget information. We recommend using the
keyring
library for persisting credentials.
import mintapi
mint = mintapi.Mint(
'[email protected]', # Email used to log in to Mint
'password', # Your password used to log in to mint
# Optional parameters
mfa_method='sms', # See MFA Methods section
# Can be 'sms' (default), 'email', or 'soft-token'.
# if mintapi detects an MFA request, it will trigger the requested method
# and prompt on the command line.
mfa_input_callback=None, # see MFA Methods section
# can be used with any mfa_method
# A callback accepting a single argument (the prompt)
# which returns the user-inputted 2FA code. By default
# the default Python `input` function is used.
mfa_token=None, # see MFA Methods section
# used with mfa_method='soft-token'
# the token that is used to generate the totp
intuit_account=None, # account name when multiple accounts are registered with this email.
headless=False, # Whether the chromedriver should work without opening a
# visible window (useful for server-side deployments)
# None will use the default account.
session_path=None, # Directory that the Chrome persistent session will be written/read from.
# To avoid the 2FA code being asked for multiple times, you can either set
# this parameter or log in by hand in Chrome under the same user this runs
# as.
imap_account=None, # account name used to log in to your IMAP server
imap_password=None, # account password used to log in to your IMAP server
imap_server=None, # IMAP server host name
imap_folder='INBOX', # IMAP folder that receives MFA email
wait_for_sync=False, # do not wait for accounts to sync
wait_for_sync_timeout=300, # number of seconds to wait for sync
use_chromedriver_on_path=False, # True will use a system provided chromedriver binary that
# is on the PATH (instead of downloading the latest version)
)
# Get basic account information
mint.get_accounts()
# Get extended account detail at the expense of speed - requires an
# additional API call for each account
mint.get_accounts(True)
# Get budget information
mint.get_budgets()
# Get transactions
mint.get_transaction_data() # as pandas dataframe
# Get transactions for a specific account
accounts = mint.get_accounts(True)
for account in accounts:
mint.get_transaction_data(id=account["id"])
# Get net worth
mint.get_net_worth()
# Get credit score
mint.get_credit_score()
# Get bills
mint.get_bills()
# Get investments (holdings and transactions)
mint.get_invests_json()
# Close session and exit cleanly from selenium/chromedriver
mint.close()
# Initiate an account refresh
mint.initiate_account_refresh()
# you can also use mintapi's login in workflow with your own selenium webdriver
# this will allow for more custom selenium driver setups
# one caveat is that it must be based on seleniumrequests currently
# seleniumrequests has most browsers already
# it also has mixins for any browsers it doesn't have so the sky is the limit!
from seleniumrequests import Firefox
mint = mintapi.Mint()
mint.driver = Firefox()
mint.status_message, mint.token = mintapi.sign_in(
email, password, mint.driver, mfa_method=None, mfa_token=None,
mfa_input_callback=None, intuit_account=None, wait_for_sync=True,
wait_for_sync_timeout=5 * 60,
imap_account=None, imap_password=None,
imap_server=None, imap_folder="INBOX",
)
# now you can do all the normal api calls
# ex:
mint.get_transaction_data()
Run it as a sub-process from your favorite language; pip install mintapi
creates a binary in your $PATH. From the command-line, the output is JSON:
usage: mintapi [-h] [--session-path [SESSION_PATH]] [--accounts] [--investment]
[--budgets | --budget_hist] [--net-worth] [--extended-accounts]
[--transactions] [--credit-score] [--credit-report]
[--exclude-inquiries] [--exclude-accounts] [--exclude-utilization]
[--start-date [START_DATE]] [--end-date [END_DATE]]
[--limit] [--include-investment] [--show-pending]
[--format] [--filename FILENAME] [--keyring] [--headless]
[--mfa-method {sms,email,soft-token}]
[--categories] [--attention]
email [password]
positional arguments:
email The e-mail address for your Mint.com account (required)
password The password for your Mint.com account (if not supplied, --keyring must be provided)
optional arguments:
-h, --help show this help message and exit
--accounts Retrieve account information (default if nothing else
is specified)
--session-path [SESSION_PATH]
Directory to save browser session, including cookies. Used to prevent repeated
MFA prompts. Defaults to $HOME/.mintapi/session. Set to None to use a temporary
profile.
--budgets Retrieve budget information for current month
--budget_hist Retrieve historical budget information (past 12 months)
--categories Retrieve your configured Mint categories
--config-file, -c File used to store arguments
--credit-score Retrieve credit score
--credit-report Retrieve full credit report & history
--exclude-inquiries Used in conjunction with --credit-report, ignores credit inquiry data.
--exclude-accounts Used in conjunction with --credit-report, ignores credit account data.
--exclude-utilization Used in conjunction with --credit-report, ignores credit utilization data.
--net-worth Retrieve net worth information
--extended-accounts Retrieve extended account information (slower, implies --accounts)
--transactions, -t Retrieve transactions
--start-date [START_DATE]
Earliest date for which to retrieve transactions.
Used with --transactions. Format: mm/dd/yy
--end-date [END_DATE]
Latest date for which to retrieve transactions.
Used with --transactions. Format: mm/dd/yy
--investments Retrieve data related to your investments, whether they be retirement or personal stock purchases
--include-investment Used with --transactions
--limit Number of records to include from the API. Default is 5000.
--show-pending Retrieve pending transactions.
Used with --transactions
--filename FILENAME, -f FILENAME
write results to file. If no file is specified, then data is written to stdout. Do not specify the file extension as it is determined based on the selection of `--format`.
--format Determines the output format of the data, either `csv` or `json`. The default value is `json`. If no `filename` is specified, then this determines the `stdout` format. Otherwise, if a `filename` is specified, then this determines the file extension.
--keyring Use OS keyring for storing password information
--headless Whether to execute chromedriver with no visible
window.
--use-chromedriver-on-path
Whether to use the chromedriver on PATH, instead of
downloading a local copy.
--mfa-method {sms,email,soft-token}
The MFA method to automate.
--mfa-token The base32 encoded MFA token.
--imap-account IMAP_ACCOUNT
--imap-password IMAP_PASSWORD
--imap-server IMAP_SERVER_HOSTNAME
--imap-folder IMAP_FOLDER
Default is INBOX
--imap-test Test access to IMAP server
--no_wait_for_sync Do not wait for accounts to sync
--wait_for_sync_timeout
Number of seconds to wait for sync (default is 300)
--attention. Get notice if there are any accounts that need attention
>>> mintapi --keyring [email protected]
[
{
"accountName": "Chase Checking",
"lastUpdatedInString": "25 minutes",
"accountType": "bank",
"currentBalance": 100.12,
...
},
...
]