Skip to content
We scan thousands of government websites to check how well they stack up on security, accessibility, and public accountability.
Python HTML CSS Dockerfile Shell
Branch: master
Clone or download
Latest commit e730a3a Oct 29, 2019


About the project

GovLens is a government transparency project developed by MuckRock and Code for Boston engineers. Our mission is to create a more open, accessible, and secure democracy through examining the technical elements of government agency websites. We use algorithms to score thousands of federal and state agencies based on their transparency, security, privacy, and accessibility. We then publish our findings and help communicate to government agencies possible improvements to their infrastructures that would better the agency as a whole.

A screenshot of what a GovLens Scorecard looks like


We get reminders all the time of how well our physical civic infrastructure is doing: Did my car hit a pothole? Are the swing sets covered in rust? It can be harder to see how well our digital civic infrastructure is holding up, however, particularly when it comes to the parts of the web that can be invisible to many people: How accessible is a site to people who rely on screen readers or who have reduced vision? Which third-party trackers have access to visitor data, and how is that data being guarded? Are government websites following basic best practices in utilizing secure connections?

While we have a National Bridge Inventory that monitors dangerous bridges and other federal agencies that monitor other core infrastructure issues, we do not have similar insights into how strong or weak much of our digital infrastructure is.

GovLens helps to provide at least the start of an answer to that, by making those oftentimes overlooked aspects of digital infrastructure more visible via public report cards for each agency in our database as well as collated data for each jurisdiction and state, letting us see which areas of the country are leading the way and which might need a little more prodding.

This is partially inspired by the work of Pulse.CIO.Gov, an official federal government website that monitored the adoption of HTTPS compliance among federal websites, as well as SecureThe.News, which did the same thing for news websites. Both of these projects brought wider visibility to the issue and provided natural and effective peer pressure for website operators to improve. Our hope is we can do the same for local government, while also compiling a rich research data set for future analysis.

Who is this site for?

This site has three core planned audiences:

  • The general public, so that they’re better educated about the state of government digital infrastructure and why it matters.
  • Government decision makers, so that they can understand why they need to invest in better adhering to web standards as well as see where their sites stand compared to their peers.
  • Local and national media outlets, so as best to reach and influence the above categories.

Getting started basics

Project goals

The goal is to create an automatically updated database that tracks, over time, how well government agencies websites at the state, local, and federal levels follow best practices when it comes to HTTPS security, mobile friendliness, reader accessibility, and other key areas.

Over time, we hope to show whether both individual agencies are improving or worsening, as well as help highlight national shifts along the metrics we monitor. Individual pages show the most recent snapshot ranking, but our API will make historical data available.

Current status

The project is currently in testing stages, as we work to both develop usable, accurate data and build a pipeline for regularly populating it. The site currently can run locally, but several of the data categories are filled with randomized testing data and any report cards generated are for demonstration purposes only. These scores do not represent actual scores for agencies.

Installation instructions

Install python3 if you haven't installed it yet. To check:

python --version
python3 --version

If it isn't installed, see Python or google it for your operating system.

Create a developer account on Github if you don't have one: Github

Fork the repository on Github, see: Fork a Repo

Clone your forked repository from the command line (this will create a GovLens directory):

git clone

Navigate to the base directory of the reposistory and prepare to install depedencies.

cd GovLens

To start, it is recommend to create a virtual environment. If you have not used virtualenv before, install it with:

# Create a virtual environment to manage dependencies
pip install virtualenv
# or
pip3 install virtualenv
# if python 2 is your default python
# Next,create a virtual environment to manage dependencies
virtualenv venv
# or
virtualenv -p python3 venv
# if python 2 is your default python
source venv/bin/activate

Now install the dependencies with pip:

# Install requirements.txt
pip install -r requirements.txt

After the dependencies have installed, we want to prepare the database.

# Perform data migrations
python migrate

Then, we need to import a CSV file containing existing agency information. Start by running a Django shell:

python shell

# From within the >>> shell
from apps.civic_pulse.utils.load_models import *

create user for the scraper. (Note that "scraper" is a bit of a misnomer, we are mostly simply calling the Lighthouse Api.)

This step is needed in order to connect the api with the scrapers. If you do not wish to do that, then this may be skipped. We need to create a dummy user for the scraper to be able to access the api. The api is part of the Django projet. Note: The scrapers live in an independent environment not neccessarily in the same server as the Django website. The scrapers read and write data to the website using api endpoints. To create a token:

  • create an admin user to be able to login to the admin portal of the website: /admin
  python createsuperuser --username admin --email
  # enter the password when prompted. It can be any password that you wish to use. 
  # It is used for login to the admin website.
  • login to the admin website and create a user for the scraper.
  • create a token for the scraper user using the following command
./ drf_create_token admin

Finally, the database is ready to go! We are now ready to run the server:

python runserver

Navigate in your browser to and you should see a list of agencies.

You can’t perform that action at this time.