the-endorser: draw out relationships between people on LinkedIn

by do son · Published July 19, 2019 · Updated November 4, 2024

The Endorser

An OSINT tool that allows you to draw out relationships between people on LinkedIn via endorsements/skills.

Check out the example (digraph), which is based on mine and my colleagues (David Prince) LinkedIn profile. By glancing at the visualization you can easily see, by the number of “arrows”, there is some sort of relationship between us and “Zoë Rose” (we all work together on the same team in this case). If I was carrying out an investigation I would focus my efforts towards her next.

Due to the way, LinkedIn’s privacy settings work this tool works best when your target is within your 3rd-degree network or higher. Using a LinkedIn Premium or Recruiter account will allow you to map targets outside of your network.

Installation

The Endorser will work on pretty much any *nix (Linux, Mac, BSD) system with Python 3.0+.

git clone https://github.com/eth0izzle/the-endorser.git
sudo pip3 install -r requirements.txt
Install Graphviz via your package manager, e.g. apt-get install graphviz or brew install graphviz
Download ChromeDriver for your platform (requires Chrome) and place in ./drivers. Alternatively, you can use PhantomJSand launch with the –driver phantomjs flag (note phantomjs is 8x slower).
Setup your LinkedIn credentials in config.yaml
python3 the-endorser.py <profile1> <profile2>

Usage

usage: python the-endorser.py https://www.linkedin.com/in/user1 https://www.linkedin.com/in/user2



Maps out relationships between peoples endorsements on LinkedIn.



positional arguments:

  profiles              Space separated list of LinkedIn profile URLs to map



optional arguments:

  -h, --help            show this help message and exit

  --config_file CONFIG_FILE

                        Specify the path of the config.yaml file (default:

                        ./the-endorser/config.yaml)

  --driver DRIVER       Selenium WebDriver to use to parse the webpages:

                        chromedriver, phantomjs (default: chromedriver)

  --output OUTPUT       Output module to visualise the relationships: digraph,

                        stdout (default: digraph)

  --log LOG             Path of log file. None for stdout. (default: None)

  --log-level LOG_LEVEL

                        Logging output level: DEBUG, INFO, WARNING, ERROR.

                        (default: INFO)

Outputs

The Endorser is “modular” in the sense that it can output and visualize the data in different ways. An output module just needs one method: def run(profiles)

Currently, there is only one output module (digraph). In the future, I plan to add modules for Maltego and Plot.ly – but feel free to get involved!

Digraph

It’s best to read this from right-to-left to identify people that have arrows from multiple profiles. Square box = skill, ellipse = person.