Existing projects

Overview of some of the projects I'm working.

Calamary - Forward & Filtering Proxy

Calamary is a squid-like forward-proxy.
Its focus is set on security filtering for HTTPS/TLS.

Since: 2023-09
State: Development
Repository: GitHub - Calamary
Documentation: ReadTheDocs - Calamary

ShieldWall - Firewall System

ShieldWall should be a system of centrally managed firewalls.
The firewalls (boxes) should only be manageable using the controller they are linked to.

Since: 2023-09
State: Development
Repositories: GitHub - ShieldWall


AnsibleGuy is a persona I created to share sub-projects related to RedHat Ansible.
I personally like to use Ansible as tool for automating IT administration as it allows one to cleanly organize complex structures/setups.

Since: 2021-09
Repositories: GitHub - AnsibleGuy
Videos: YouTube


Helping to optimize grow environments.
I used this project to learn programming.

Lifetime: 2020-01 - 2022-01
State: Deprecated
Repository: GitHub - GrowAutomation
Documentation: ReadTheDocs

Decentralized Tor alternative

Maybe based on UDP-QUIC to speed it up? Haven't put much though into it yet.

Lifetime: 2023-01 - ?
State: Planning
Repository: GitHub - ShadowRealm

Project backlog & ideas

Overview of some project ideas i would like to tackle in the future.
Feel free to contact me if you would like to collaborate on any of these projects!
It would also be nice to know if you encountered any existing OpenSource projects that focus on a topic listed here.

Digital-First schooling

I'm from Austria πŸ‡¦πŸ‡Ή. Like many others I've encountered many frustrating problems while I attended our schools.
Many times the teachers are the worst part. The either don't care about the students or are not made out to help others learn.
Some of them even dislike single students personally and make their life harder than is should be.
One might say: "This is life. Go deal with it.." - but I don't see it that way.
I strongly believe that we could improve the experience of our students. They might even start to like some discipline they hated before.

In my opinion we should focus on a Digital-First school-system!
If you have experienced how nice it is to learn with tools like YouTube tutorials, Apps like Duolingo or Platforms like Brilliant - you might acknowledge how they surpass most of our school classes.
I would like to see a centralized portal that provides the students with video content that covers the curriculum.
Students will get specialized tablets that can only access the provided/needed 1-3 apps. (+ learning music)
They should watch these videos in the cassroom. If they have questions - they can talk to the teacher. In that way each student can learn at the pace they like/need. If they don't understand the lesser at the first time - they can watch it repeatedly.

This could also help improve out school system itself:

  • teachers don't need to repeat themselves each year
  • videos can be optimized and extended every year (a public feedback system like in code-repositories may make sense)
  • students will have equal opportunities
  • the central system could include forums which can be used by students and teachers to ask further questions and search for existing explainations
  • teacher can focus on their individual duties and/or improve the system

Digital Self Service for Restaurants

Web service to enable restaurants to deploy self-service tablets to its tables.
The orders could be shown on a screen in the service area.
To give a better overview of the orders and current status of any table - the Web service should allow the use of maps. The customer or maintainer would add a map of the restaurants rooms and create/link all the existing tables on it.
Maybe toilets and exits could also be added to the map (viewable by the customer) to add useful information.

A table object should be in one of multiple states. Per example: cash, waiting for drink, waiting for meal (support multiple ones), eating, after some idle time a 'check on' notifier
The status of the current order could be updated via 2 click/taps (table + pre-defined status) so the customer would see what's going on.
I think this would lower the stress of personal and add transparency for the customers. Win-Win.
The customer could also get creative and box the tablets inside the table, in boxes or fixate them using a cable.

It acting as a billing system might be necessary. This might create a lot more complexity. Also printing barcodes might bring some problems with it.
Maybe at first an API interface for transferring the order-data would be an option. But that does not feel like a clean solution.

To setup such a system one would need some experience and insight in how it works.
This is where maintainers could be used.
If the solution is hosted - they could be encouraged by giving them a cut of the monthly per-tablet license-costs.
If the solution is self-hosted - they could bill the custom per setup.

WireGuard extension

Extend the WireGuard-Go implementation to:

  • support Multi-Factor-Authentication (LDAP, TOTP, ...)
  • option to Tunnel it over TCP+TLS/HTTPS to get around firewalls
  • option to run it as unprivileged user (client has a SOCKS5 entrypoint instead of modifying routes)

JA3 Database

Create & maintain a public JA3 SSL-Fingerprint database.
Allow admins to register an account and contribute the JA3 hashes they encountered.
How the contribution could work:

  • process each request-log locally
  • check it against the cached database
  • if the hash is unknown or not marked as 'finalized' it can be pushed to the database
  • the push would at least need this information: JA3-hash (algorithm must be fixed), user-agent or similar client-/protocol-information, protocol/application used
  • the contributions per account and day are very limited as we do not expect that many new hashes appear
  • an account can gain or lose reputation
  • if a hash is pushed by multiple accounts it gains reputation
  • users can filter by reputation and contributor accounts when downloading the database

Real Sign

Decentralized & privacy retaining solution to identify real media (anti AI-deepfake).
Haven't put much though into it yet.