sooryathejas/METATRON

in llm.py

```
OLLAMA_BASE_URL = os.getenv("OLLAMA_BASE_URL", "http://localhost:11434").rstrip("/")
OLLAMA_URL = f"{OLLAMA_BASE_URL}/api/chat"
```

then use: `export OLLAMA_BASE_URL="http://10.32.31.40:11434"`

Hey, just wanted to give you a quick update — everything is up and running perfectly.

I set it up on Kali, and the whole stack is working 100%:

* MariaDB initialized and handled data correctly
* Ollama running smoothly with the model loaded
* METATRON executing without issues and interacting with the model as expected

End-to-end flow is solid — scans, processing, and responses are all working cleanly.

Honestly, it turned out really solid. Nice work on this.

METATRON works on any Debian-based distro — Kali
included. No special changes needed.

Install dependencies the same way:
pip install -r requirements.txt
sudo apt install nmap whois whatweb curl dnsutils nikto

Only thing to note is Ollama needs to be installed separately since it's not in apt by default:(depends on your kernel):
curl -fsSL https://ollama.com/install.sh | sh

Should work out of the box. Let me know if you
hit any issues.

> > sounds cool to me. i was initially planning to do that but i didnt have the time right now, will try to implement in future update.
>
> Do you plan to add Kali Linux compatibility with that update?
>It is compatible with kali linux and other debian based linux.

> sounds cool to me. i was initially planning to do that but i didnt have the time right now, will try to implement in future update.

Do you plan to add Kali Linux compatibility with that update?

Hi Soorya,*
I have added all the tools that hexstrike uses to METATRON, I am going to be running some tests on it to see how it performs.

Thankyou for your proposal,
I will think about it and let you know

*Subject: Business Proposal: Advanced & Secured Build of METATRON*

*Dear Soorya,*

Thank you for your response on GitHub. Since you don't use X, I am reaching
out here to discuss the commercial potential of the enhanced version of
METATRON I’ve developed.

As you know, I have identified several critical vulnerabilities (RCE, SSRF,
etc.). My private build not only patches these flaws but also implements a
high-performance *multi-threading* engine, capable of scanning thousands of
targets in record time without compromising system security.

I am interested in offering you this proprietary build under a *commercial
license* or a *complete buy-out* of the code, so it can be integrated into
a "Pro" version of the project.

I have attached a brief performance summary of my latest scan (1,000
targets) to demonstrate the efficiency of this build.

Looking forward to hearing your thoughts on a potential deal.

*Best regards,* *ZAKARIA*



‫في الخميس، 9 أبريل 2026 في ...

Hi zakaria, I don't use X (twitter) I will share my telegram in profile.

> You can. Edit your llm.py file to point to your external ollama machine

It should be configurable a-la an env var instead of editing the files. (Because then we have a divergence from the git tree)

You can. Edit your llm.py file to point to your external ollama machine

sounds cool to me. i was initially planning to do that but i didnt have the time right now, will try to implement in future update.