Ctrlk
Changelog NotificationsAPIAikido login
For the complete documentation index, see llms.txt. This page is also available as Markdown.

Python (Flask)

This guide will walk you through installing and setting up Zen Firewall by Aikido for your application. Follow the steps below to protect your application.

If you encounter any issues or problems, don't hesitate reach out on support chat or Github issues

https://github.com/AikidoSec/firewall-python

Requirements

Installation & Configuration

1

Install Zen Firewall by Aikido

Install Zen with pip:

pip install aikido_zen

Set your token environment variable:

export AIKIDO_TOKEN=AIK_RUNTIME_YOUR_TOKEN_HERE

Enable Zen at the top of your entrypoint, before other imports when possible:

import aikido_zen

aikido_zen.protect()

Your app.py should initialize Zen before other imports:

import aikido_zen
aikido_zen.protect()

from flask import Flask
2

Enable Request Blocking and User Identification

Use this middleware to enable rate limiting, user identification, and blocking features.

Zen Firewall does not require this middleware to block attacks. Core attack protection works without it. The middleware provides additional request context that Zen uses for protections such as rate limiting, user blocking, bot blocking, country blocking, and threat actor blocking.

Adapt the example to fit how your application identifies users and handles requests.

Add the Flask middleware:

from aikido_zen.middleware import AikidoFlaskMiddleware

app = Flask(__name__)
app.wsgi_app = AikidoFlaskMiddleware(app.wsgi_app)
# Put auth middleware before this for user-aware rules.

Set the user in your auth flow:

from aikido_zen import set_user

set_user({"id": "123", "name": "John Doe"})
3

Start Zen Firewall in dry / detection-only mode

AIKIDO_BLOCK=false AIKIDO_TOKEN=YOUR_SECRET_TOKEN flask run

Set the token as an environment variable so the Aikido Zen agent can pick it up. If you don't have a token yet, follow instructions here.

AIKIDO_TOKEN=AIK_RUNTIME_

We recommend to start your app in dry mode to ensure it works as expected without blocking any requests. We advise running Zen Firewall in staging for two weeks to avoid false positives.

AIKIDO_BLOCK=false

You can use AIKIDO_DEBUG=true to enable debug mode for more detailed information about what the agent is doing. For more information about your environment variables: Configuration via Environment Variables

4

Test your app

Browse to your application and perform a couple of actions or open a couple of pages. Zen will automatically discover the routes in your application.

Zen sends data back to Aikido every 10 minutes

You can verify a working agent by looking at the following pages of your Zen application:

  • Events: Should show an "Application started" event.

  • Routes: After some time your application routes will start showing here with the method, route and requests.

  • Instances: Should show the number of active instances for your application where Zen is installed.

5

Setup rate limiting in the dashboard

When you've added the Zen Firewall middleware you can test protecting a route from brute force attacks, you do this by setting up rate limit in the Aikido Dashboard:

  1. Click on the created app.

  2. Go to the Routes tab.

  3. Find the route you would like to limit and click Setup rate limiting.

  4. Follow the instructions to configure the rate limit (e.g., 5 requests per minute).

API route management interface showing authentication routes with protection and rate limiting options.
Set rate limiting for POST /auth/login to 5 requests per minute.

Verify Rate Limiting

Start your app and try to access the route you've rate limited 5 times within a minute. After the fifth attempt, you should receive a rate limit error:

You are rate limited by Aikido firewall. (Your IP: 1.2.3.4)
6

Next steps

Congrats you've successfully installed Zen Firewall. If you encountered any problems, have concerns or feature requests, don't hesitate to reach out to support.

You can now go and explore the many features that Zen Firewall provides:

Additional information:

Last updated

Was this helpful?