spyboy (@itisspyboy) header
spyboy (@itisspyboy) profile picture

VAPT Analyst , Penetration Testing, Vulnerability Assessment, Cyber Security Researcher https://nitter.gq/t.co/R9S5YqQnH1

Joined February 2012
Photo by spyboy (@itisspyboy)Photo by spyboy (@itisspyboy)Photo by spyboy (@itisspyboy)Photo by spyboy (@itisspyboy)Photo by spyboy (@itisspyboy)Photo by spyboy (@itisspyboy)Photo by spyboy (@itisspyboy)Photo by spyboy (@itisspyboy)Photo by spyboy (@itisspyboy)Photo by spyboy (@itisspyboy)
spyboy retweeted
Europe came to lecture India on climate responsibility... The unexpected twist? Piyush Goyal came prepared with facts, figures & reality check to slap them hard 😂🔥
403
8,822
221
38,305
750,402
spyboy retweeted
💯
95
4,398
111
23,035
560,262
spyboy retweeted
THIS IS A BIG PROBLEM FOR US! Everyone is recording videos of themselves doing regular tasks and getting up to ₹250/hour. But this has a Dark Truth. Companies collect this Egocentric data to sell to AI companies, so you can be replaced with Humanoids very soon. And if you still don't understand what's happening or think this is nothing, watch our video fully to know everything: youtu.be/897AQn6XqmQ?si=gE7t… What do you think about these apps and services? Tell us!
23
50
1
1,070
122,834
🚨 BREAKING: Google DeepMind just mapped the attack surface that nobody in AI is talking about. Websites can already detect when an AI agent visits and serve it completely different content than humans see. > Hidden instructions in HTML. > Malicious commands in image pixels. > Jailbreaks embedded in PDFs. Your AI agent is being manipulated right now and you can't see it happening. The study is the largest empirical measurement of AI manipulation ever conducted. 502 real participants across 8 countries. 23 different attack types. Frontier models including GPT-4o, Claude, and Gemini. The core finding is not that manipulation is theoretically possible it is that manipulation is already happening at scale and the defenses that exist today fail in ways that are both predictable and invisible to the humans who deployed the agents. Google DeepMind built a taxonomy of every known attack vector, tested them systematically, and measured exactly how often they work. The results should alarm everyone building agentic systems. The attack surface is larger than anyone has publicly acknowledged. Prompt injection where malicious instructions hidden in web content hijack an agent's behavior works through at least a dozen distinct channels. Text hidden in HTML comments that humans never see but agents read and follow. Instructions embedded in image metadata. Commands encoded in the pixels of images using steganography, invisible to human eyes but readable by vision-capable models. Malicious content in PDFs that appears as normal document text to the agent but contains override instructions. QR codes that redirect agents to attacker-controlled content. Indirect injection through search results, calendar invites, email bodies, and API responses any data source the agent consumes becomes a potential attack vector. The detection asymmetry is the finding that closes the escape hatch. Websites can already fingerprint AI agents with high reliability using timing analysis, behavioral patterns, and user-agent strings. This means the attack can be conditional: serve normal content to humans, serve manipulated content to agents. A user who asks their AI agent to book a flight, research a product, or summarize a document has no way to verify that the content the agent received matches what a human would see. The agent cannot tell the user it was served different content. It does not know. It processes whatever it receives and acts accordingly. The attack categories and what they enable: → Direct prompt injection: malicious instructions in any text the agent reads overrides goals, exfiltrates data, triggers unintended actions → Indirect injection via web content: hidden HTML, CSS visibility tricks, white text on white backgrounds invisible to humans, consumed by agents → Multimodal injection: commands in image pixels via steganography, instructions in image alt-text and metadata → Document injection: PDF content, spreadsheet cells, presentation speaker notes every file format is a potential vector → Environment manipulation: fake UI elements rendered only for agent vision models, misleading CAPTCHA-style challenges → Jailbreak embedding: safety bypass instructions hidden inside otherwise legitimate-looking content → Memory poisoning: injecting false information into agent memory systems that persists across sessions → Goal hijacking: gradual instruction drift across multiple interactions that redirects agent objectives without triggering safety filters → Exfiltration attacks: agents tricked into sending user data to attacker-controlled endpoints via legitimate-looking API calls → Cross-agent injection: compromised agents injecting malicious instructions into other agents in multi-agent pipelines The defense landscape is the most sobering part of the report. Input sanitization cleaning content before the agent processes it fails because the attack surface is too large and too varied. You cannot sanitize image pixels. You cannot reliably detect steganographic content at inference time. Prompt-level defenses that tell agents to ignore suspicious instructions fail because the injected content is designed to look legitimate. Sandboxing reduces the blast radius but does not prevent the injection itself. Human oversight the most commonly cited mitigation fails at the scale and speed at which agentic systems operate. A user who deploys an agent to browse 50 websites and summarize findings cannot review every page the agent visited for hidden instructions. The multi-agent cascade risk is where this becomes a systemic problem. In a pipeline where Agent A retrieves web content, Agent B processes it, and Agent C executes actions, a successful injection into Agent A's data feed propagates through the entire system. Agent B has no reason to distrust content that came from Agent A. Agent C has no reason to distrust instructions that came from Agent B. The injected command travels through the pipeline with the same trust level as legitimate instructions. Google DeepMind documents this explicitly: the attack does not need to compromise the model. It needs to compromise the data the model consumes. Every agentic system that reads external content is one carefully crafted webpage away from executing attacker instructions. The agents are already deployed. The attack infrastructure is already being built. The defenses are not ready.
36
129
15
332
42,342
Free Mail Forwarding forward.haltman.io/
8
37
4,870
Google CEO Sundar Pichai on current frontier model's ability to break the security of almost all current software. "These models are definitely, like really gonna break pretty much all software out there, maybe already, we don't know."
197
432
35
2,948
947,498
spyboy retweeted
INSTEAD OF WATCHING NETFLIX TONIGHT. Spend 1 hour building this. Obsidian + Claude Code = your own personal JARVIS. A second brain that captures everything, connects every idea, and thinks alongside you using the most powerful AI model available. Takes 1 hour to set up. Works while you sleep. The people who build this tonight will never work the same way again. The people who skip it will still be taking scattered notes and losing their best ideas next year wondering why they cannot think clearly. Your call.
85
607
26
5,572
478,703
2004 was a good year, but your Gmail address doesn't need to be stuck in it. To say goodbye to [email protected] or [email protected] (or whatever you were into at the time), go to your Google Account settings and choose any name available. You'll keep your old username and you can sign in with both.
1,071
3,014
679
39,830
4,944,634
It is quite well known that Protonmail has helped trace multiple activists over the last few years who trusted their word. You would be shocked to hear their own response when confronted about their bold claims. Proton ≠ Complete Privacy
This quoted post is unavailable.
69
1,115
39
7,137
288,416
It only took 14 years… but it’s finally here 😊 Meet NexPhone — a smartphone built to run Android, launch Linux (Debian) on demand, and dual-boot Windows 11. My 14-year founder story: nexphone.com/blog/the-tale-o… If you want to support what we’re building, a repost helps a lot.
439
2,077
233
10,214
648,886
Truth!🤓
290
841
44
10,490
228,350
BillionMail is an open source, self hosted email marketing platform that puts devs in control. The latest update brings two powerful features to boost deliverability and engagement 👇 🌐 Dedicated IPs per domain. ✨ Spintax Support.
6
76
3
722
88,404
Netflix-like experience, but torrents.
104
665
46
12,037
841,100
SIM Swap Scam: A SIM swap scam is when someone pretends to be you and gets your phone number from your phone company. Then they can use your phone number to get into your online accounts, like your bank, email, or social media. They can take your money, data, or identity. In 2021, a woman from New York lost $22,000 after a hacker swapped her SIM card and got into her online banking account. The hacker also locked her out of her email and social media accounts, and tried to get more money from her contacts How a SIM swap scam works: 1. The scammer finds out your personal information from public websites, social media, social engineering... 2. The scammer calls your phone company and acts like you. They ask to move your phone number to a new SIM card that they have. 3. The phone company asks some questions to check your identity. The scammer answers them using the information they learned about you. 4. The phone company moves your phone number to the scammer’s SIM card. You lose service on your phone and the scammer gets your calls, texts, and codes. 5. The scammer uses your phone number to sign into your online accounts, such as your bank, email, or social media. They can steal your money, data, or identity. How to avoid SIM swap scam: 1. Don't share your personal information, like your phone number, address, or social security number, on public websites or social media. 2. Make a PIN or password for your mobile account that only you know. Don’t use things that are easy to guess, like your birthday or address. 3. Don't send your OTP (One-Time Password) to anyone. 4. Use apps or physical keys instead of texts for your online accounts that need two steps to log in. Texts can be stolen by someone who has your phone number. 5. Check your phone and bank statements often for any strange activity.
12
74
1
241
23,738
face swap and face manipulation toolkit
11
193
3
2,445
217,979
Facad1ng: An open-source URL masking tool designed to help you Hide Phishing URLs and make them look legit using social engineering techniques GitHub: github.com/spyboy-production…
4
71
332
22,927
track gps location, capture device info and pics
1
2
20
3,043