BackBox.org

A Free Open Source Community Project

[Raffaele Forte @raffaele_forte]

14 voli minacciati in un giorno, circa 6.000 scuole italiane in una mattina, nemmeno una bomba. Con BackBox AI abbiamo ricostruito le ondate di falsi allarmi in Europa su 139 fonti aperte: sei firme, nessuna attendibile.

Articolo e dashboard: https://t.co/azz9Va8OLu

[BackBox Labs @backbox_labs]

14 flights threatened in one day, some 6,000 Italian schools in one morning, not a single bomb. With BackBox AI we reconstructed Europe's hoax bomb-threat waves from 139 open sources: six signatures, none reliable.

Article and live dashboard: https://t.co/cA3uFdau3v

[Raffaele Forte @raffaele_forte]

Due notti, lo stesso copione: refrigerazione manomessa in centinaia di supermercati francesi (11-12 ottobre), poi in circa 170 siti europei, Italia compresa (1-2 novembre). Nessun malware, nessuna rivendicazione.

L'indagine OSINT con BackBox AI: https://t.co/TV0gcqK6Z1

[BackBox Labs @backbox_labs]

Two nights, one playbook: refrigeration controllers tampered with in hundreds of French supermarkets (11-12 Oct), then in about 170 sites across Europe (1-2 Nov). No malware, no claim, no attribution.

Our OSINT investigation with BackBox AI: https://t.co/ooT0C0c2bD

[Raffaele Forte @raffaele_forte]

OSINT su una crisi in corso: come @backbox_labs ha monitorato le proteste di ottobre in Francia e Italia con BackBox AI Threat Hunter.

Metodo, risultati, limiti delle fonti aperte e dashboard dell’indagine.

https://t.co/R5rRwaAv0u

[BackBox Labs @backbox_labs]

October's protests in France and Italy, followed with BackBox AI on open sources only. No common command observed, mostly domestic disinformation, predictions checked against events.

Article: https://t.co/aIPBxNT1Hx
Dashboard: https://t.co/AvoPVb6Zig

[BackBox Labs @backbox_labs]

BackBox AI built a complete Chrome 152 exploit chain against @Google's v8CTF target, from a web page to RCE and a file read.

Three V8 bugs, a sandbox escape, WebAssembly and ROP.

We break down how it works and what the patch gap means for defenders.

https://t.co/6DNdJDqS4i

[BackBox Labs @backbox_labs]

We put BackBox AI to the test on @hackthebox_eu Season 11.

13 machines. Easy to Insane. How far could it get on its own?

We share what our agent achieved, where it struggled, and what we learned about autonomous offensive security.

https://t.co/72a37BVMtO

[BackBox Labs @backbox_labs]

0-day in @nopCommerce

BackBox AI discovered a stored XSS that bypasses both nopCommerce's HTML sanitizer and a fronting WAF using an SVG `animate onbegin` vector.

The vulnerability remains unpatched through version 4.90.6.

Full advisory: https://t.co/yUtgXUS7gA

[BackBox Labs @backbox_labs]

The future of AI-powered cybersecurity isn't about owning the best model. It's about building the best AI harness.

Read why: https://t.co/QN8PBD1Rc5

[BackBox Labs @backbox_labs]

New @netbird 0day discovered!

Our research found a local privilege escalation affecting NetBird Linux clients (≤ 0.75.0), allowing local users to gain root via IPC design flaws.

Technical analysis: https://t.co/hKxdHyGcso

[Raffaele Forte @raffaele_forte]

Discovered a zero-day vulnerability affecting a modern zero-trust mesh networking implementation. Responsible disclosure is underway. Technical details will be shared soon.

https://t.co/KqIE6hlXUe

[Raffaele Forte @raffaele_forte]

OpenAI's incident involving Hugging Face shows that AI agents can discover vulnerabilities. This is not new.

Open models can be equally effective in offensive security with the right architecture and tools.

The future is secure agent design.

[BackBox Labs @backbox_labs]

This summer, leave your laptop behind.

BackBox AI is available on mobile. All you need is a
smartphone and an internet connection.

Try it:
https://t.co/s2LorjtgRb

[BackBox Labs @backbox_labs]

BackBox AI analyzed a live Android RAT from a single APK, extracting the implant, uncovering obfuscated C2 infrastructure, and producing evidence-backed attribution through adaptive AI-driven reasoning.

Read the full technical analysis:
https://t.co/K2JVvPV4ui

[BackBox Labs @backbox_labs]

La sovranità tecnologica nell'era dell'AI non è un tema futuro. È una scelta di oggi.

Modelli aperti, controllo dei dati e libertà di scelta: la nostra direzione.

https://t.co/7gHuaUEaM1

[BackBox Labs @backbox_labs]

New security advisory: Bettercap LPE affecting Debian, Ubuntu, Arch and derivatives.

Not a Bettercap code flaw, but an insecure service packaging issue.

https://t.co/fVGyZr2KZ7

[BackBox Labs @backbox_labs]

New security advisory: BackBox AI found a memory exhaustion DoS in Binwalk's CSMAN parser. A crafted file could trigger unbounded decompression during a normal scan. The issue was responsibly disclosed and fixed upstream.

https://t.co/Ab6le5Dgq9

[BackBox Labs @backbox_labs]

Security Advisory: Header Trust Issues in SearXNG

BackBox AI identified two header trust vulnerabilities in SearXNG, including a Host header injection that could lead to open redirect and cache poisoning, and an X-Forwarded-For based rate limiter bypass.

https://t.co/rxpkk90faE

[Raffaele Forte @raffaele_forte]

Hey pentesting distro maintainers (@kalilinux, @ParrotSec and others), DM us.

We identified a default package inherited from Ubuntu/Debian/Arch that allows local privilege escalation.

Details available privately for coordinated disclosure.