Twitxy Saved

MAX

Verified account This account is verified. Learn more

@MAXdeg0

Red team by day. Prompt-injecting by night. Breaking AI responsibly.™
Joined October 2023
277 Following 18K Followers 102K Posts

Latest posts 5

Pinned post

MAX Verified account @MAXdeg0
THIS IS F**KING DANGEROUS.

SOMEONE JUST UNLEASHED AN OPEN SOURCE AI HACKING ARMY TOOL
THAT CAN LAUNCH AUTONOMOUS PENTESTING OPERATIONS.

It’s called SAPPER.

One AI commander. An entire swarm of AI agents. Real hacking tools running behind the scenes.

Here’s what makes it stand out:

→ One AI commander plans and coordinates the operation

→ Multiple AI agents work together to execute tasks

→ Runs real tools like Nmap and Nuclei

→ Built in hacking knowledge base for the agents

→ Integrates with Claude Code and Codex

→ Runs inside a Kali Linux environment

It reportedly competed against 634 teams in a Tencent attack and defense competition and operated for 8 hours without human intervention.

Fully open source.

one commander —> multiple ai hackers—> an entire automated red team

Only use it in authorized labs and CTF environments.

REPO BELOW
0:29
138 replies 428 reposts 3.1K likes 349K views View on X
MAX Verified account @MAXdeg0
The next AI race may come down to context rather than raw model capability.

Models can already draft, summarize, and answer questions well. Yet before an AI can help with anything meaningful, people still do a surprising amount of prep work: explaining the project, digging up old history, clarifying what changed, and deciding which tool should handle the task.

That is the friction J.Doe is trying to rethink. Its approach connects real-world experiences with digital context, then uses that understanding to suggest a possible next step.

The ambition is personal intelligence: an AI that becomes more useful to one particular person over time, instead of simply producing a slightly better generic answer for everyone.

That is a harder problem than making a chatbot sound smarter. It may also be the more important one.
J.doe Verified account @meetJdoe 1h
Life doesn’t happen in a prompt box. Neither should your AI.

Today, we’re opening the Private Beta for J.Doe—personal intelligence that grows with you.

Most AI still waits behind an empty input box, asking you to remember the conversation, reconstruct the background, and write the prompt yourself.

J.Doe starts earlier:
• Connects real-world conversations with what’s on your screen
• Anticipates a concrete next step before you ask
• Lets you decide with a single gesture—then hands approved work to agents like Codex and Claude Code while you get back to your life

Your mind was made for more.
View quoted post on X
8 replies 1 reposts 16 likes 928 views View on X
MAX Verified account @MAXdeg0
SOMEONE JUST LEAKED THE CONTENT WORKFLOW BEHIND A 3M VIEW POST.

he didn’t film a single video.

he used claude fable 5.2 to break down a viral slideshow, recreate it for his own app, generate 5 variations and schedule a full week of content using postiz.

even better, he shared every prompt.

Full playbook below
0:10
Inoms Verified account @Inomsxbt 3h
View quoted post on X
13 replies 5 reposts 26 likes 1.2K views View on X
MAX Verified account @MAXdeg0
this is the most DANGEROUS thing on internet today.

a chinese developer just built an AI hacking platform that brings an entire red team operation into one place.

it’s called full chain sentinel.

→ automated web and API penetration testing

→ android and ios security testing

→ reconnaissance and attack surface discovery

→ vulnerability detection and exploit validation

→ MITM traffic interception and analysis

→ AI-powered security assessments and reporting

it combines multiple AI agents with a kali environment to automate security testing from reconnaissance to reporting.

one platform. an entire AI powered hacking toolkit.

Repo below
0:24
19 replies 9 reposts 93 likes 5.8K views View on X
MAX Verified account @MAXdeg0
finally, a better way to manage spending across your ai agents.

@nasikolabs just launched its open runtime platform, giving teams visibility and control over their entire agent infrastructure.

→ discover every agent running across the company

→ track spending by team, model and agent

→ monitor costs across multiple providers

→ get alerts when budgets reach 80%

→ automatically pause agents when they hit spending limits

deployment takes just one command. no sdk, wrappers or migration required.

if you’re running ai agents at scale, keeping track of costs shouldn’t be another headache.

check it out: bit.ly/nasikolaunch
www.nasiko.com Nasiko | Open Runtime for Agents Nasiko helps enterprises account for what their AI agents cost, can access, and do, across every model and provider.
0:58
Nasiko @nasikolabs 20h
Nasiko - Open Runtime platform for managing AI agents.
See every agent. Cut token spend by up to 50%. Route to the right model with the agent gateway. Govern tool access. Set permissions and budgets per agent, per team. One command. Same workflow.
Open Runtime for agents starts now. First, every token. Next, every action!
Start Free: buff.ly/k79CiWW

#ai #tokenops #openruntime #agents
View quoted post on X