Skip to main content
Version: 2.0.0

Linux Installation

Keploy can be installed in two ways:

  1. One-Click Install.
  2. Manual Install

One click install Keploy.

 curl --silent -O -L && source

Manual Install

There are two ways to use Keploy eBPF in linux, you can use either use:

  1. Natively in Linux.
  2. Through Using Docker.

Linux Native

Download the Keploy Binary

curl --silent --location "" | tar xz -C /tmp

sudo mkdir -p /usr/local/bin && sudo mv /tmp/keploy /usr/local/bin && keploy
ARM Architecture
curl --silent --location "" | tar xz -C /tmp

sudo mkdir -p /usr/local/bin && sudo mv /tmp/keploy /usr/local/bin && keploy

Run the Record Mode

Run this command on your terminal to start the recording of API calls:-

sudo -E keploy record -c "CMD_TO_RUN_APP"

Make API Calls using Hoppscotch, Postman or cURL command.

Keploy with capture the API calls you have made to generate the test-suites which will contain the testcases and data mocks into YAML format.

Run the Test Mode

Run this command on your terminal to run the testcases and generate the test coverage report:-

sudo -E keploy test -c "CMD_TO_RUN_APP" --delay 10

Voilà! 🧑🏻‍💻 We have the server running!

Using Docker

We need to create a custom network for Keploy since we are using the Docker, therefore application container would require docker network to act as the bridge between them.

If you're using a docker-compose network, replace keploy-network with your app's docker_compose_network_name below.

docker network create keploy-network

Capture the Testcases

Now, we will record the testcases.

keploy record -c "docker run -p <appPort>:<hostPort> --name <containerName> --network keploy-network --rm <applicationImage>" --containerName "<containerName>" --delay 10

Run the Testcases

Now, we will test the testcases.

keploy test -c "docker run -p <appPort>:<hostPort> --name <containerName> --network keploy-network --rm <applicationImage>" --containerName "<containerName>" --delay 20

CMD_to_run_user_container is the docker command to run the application. If you are using docker compose command to start the application, --containerName is required.

Voilà! 🧑🏻‍💻 We have the server running!

You'll be able to see the test-cases that ran with the results report on the console as well locally in the testReport directory.


  1. delay is required while using Test Mode.
  2. containerName is optional if you are using Docker run command, as the Container name must be present within the command itself.