KismetUIPlugin

Pretty useful graphs for Kismet like below:

Installing

Tested on Kali Linux release 2023.2 (ARM, Pi4)

Install Kismet then:

sudo apt-get install python3-pandas python3-pandas-lib libopenblas-dev

pip3 install -r requirements.txt

chmod u+x start.sh

chmod u+x stop.sh

You may want to follow the instructions here to install kismet as a managed service (note that you may need to set the user and group of the modified system service to your usual username and group, not kismet!). If you do this, remove the respective lines for kismet in start.sh and stop.sh.

I found that using the remote capture tool kismet_cap_linux_wifi was more reliable for running over extended periods than adding local capture interfaces to the kismet GUI. See start.sh which includes an example that you will need to edit to meet your needs.

By default this visualisation server runs on port 8050, change the line at the bottom of the python code if needed.

Using

Edit start.sh to meet your interface needs then:

./start.sh
./stop.sh

Put something like this in your crontab file to autostart on boot then also restart every 24 hours and manage database growth.

User interface

You can interact with the graph (important as information may be hiding off the screen) as follows:

Mouse	Feature
Wheel	Zoom in and out
Left button	Drag whole graph or specific nodes
Hover	Node and edge information

You can interact with the menu as follows:

Label	Explanation
Channel	Select all or a specific channel seen, this list is updated dynamically by refreshing your browser (CTRL F5)
Graph type	Choose what to graph
Kismet credentials	username:password e.g. me:12345
Kismet URI	Kismet API URI e.g. 192.168.1.50:2501
Rewind timeframe (s)	How long back you will start to look, bigger values mean bigger graphs, and too small will yield nothing
Refresh	Refresh the graph based on what you have
Reset	Reset the graph back to defaults

Brucie Bonus, real time Probe alerts and better visibility

You can run start_realtime.sh as root and then get MQTT alerts when a probe comes in based on what is in what is in process_real_time_probes.example.cfg.

Also The 'Probes as CSV' button on the UI can be used to dowload a file, which when processed with probe_viewer.py gives you this:

Enjoy