[thecrowsec]

Hey HF,

I'm currently reviving an older social engineering framework called Dragnet, originally presented at DEF CON. The goal is to modernize and upgrade it into a new tool: DragPhish (aka Dragnet v2.0)

⚙️ What’s DragPhish?

A revamped version of the original Dragnet social engineering automation tool, now being updated with:

✅ Bug fixes
✅ Dependency upgrades
✅ Refactored modules
✅ Planning full ML + AI integration
✅ New OSINT data feeds
✅ Smarter campaign recommendation engine

The vision is to turn DragPhish into a modern intelligent phishing toolkit — one that learns from past campaigns, adapts strategies, and auto-curates high-conversion payloads and lures.

🤖 What I'm Looking For

I'm seeking help from folks with experience in:

🧠 Machine Learning / NLP (TensorFlow, PyTorch, LLMs, scikit-learn)
🛰️ OSINT Integration (public datasets, LinkedIn scraping, social APIs, etc.)
🧰 Python, Node.js, Firebase, Asterisk, Vue.js
🤓 Anyone passionate about automating SE/red team tasks

🧪 Sample Tasks

Integrate NLP to analyze and auto-generate phishing emails based on target profiles
Feed DragPhish conversion history into a small ML model to recommend pretexts
Add better OSINT data enrichment modules (LinkedIn, HaveIBeenPwned, DNS scraping)
Automate infrastructure deployment (domain + landing page + mail spoof + VoIP call)

🚧 Repo / Access / Contributions

Currently private, but I’ll open early alpha to trusted contributors
DM me here if you’re interested — happy to share a walk-through of current progress
Will credit all contributors in the repo + write-up when public release drops

🗨️ Why Now?

Dragnet was great in its time but outdated. SE campaigns today need real-time intelligence, adaptive methods, and smarter automation. AI makes that possible. Let’s build something that red teams (and even blue teams) can actually use.

Drop a reply or shoot a DM if you're interested in collaborating 👇
Let’s bring Dragnet back — smarter and meaner.

[skull]

In general, LLMs are capable of this task, however any hosted LLM will block this usage.

Are you planning to run a local LLM for this? In order to bypass the restrictions set on any hosted LLM?

[thecrowsec]

In general, LLMs are capable of this task, however any hosted LLM will block this usage.

Are you planning to run a local LLM for this? In order to bypass the restrictions set on any hosted LLM?

Yes im looking to try it locally and on server, I just dont have good expertise in LLM and stuff, Ive heard of lama and stuff and Im looking to try it