Andy
@_andydeng
Followers
167
Following
146
Media
26
Statuses
198
Code & Design, building @AgentLISA_ai and @codenut_ai / Lang: en, de, zh
Singapore
Joined May 2022
Security isn’t optional, it's autonomous. 🛡️ Introducing @AgentLISA_ai now live on Questflow. LISA is the world’s first Agentic Security Operating System for Web3, purpose-built to secure smart contracts and agentic autonomous payments with LLM based multi-agents solutions.
3
6
15
🔥
Earlier this year we announced x402. x402 enables AI agents to pay for content and services as well as execute transactions using stablecoins. In the last 30 days we’ve seen over $200M in annualized transactions across both humans and AI agents. 📈
0
0
0
Nicely put. It’s like doctors can never earn reputation if they have only been working for a year or two. You got to see thousands of weird cases firsthand before becoming a battle tested expert everyone would respect and trust.
Proposing a new metric for AI vulnerability scanning tools: Total Token Consumption If your AI scanning tool claims great results after consuming only 10 million tokens, those results are likely fragile. Consumed 10 billion tokens? Now we're talking about something solid. This
1
4
4
Just went through the entire code base of x402 v2. If you want to save some time, this is a detailed summary worth reading. Those (code) comparisons with v1 provide a practical way for understanding all the important changes and improvements
0
1
3
3 highlights for me: 1. Modular paywall package - this makes the core more lightweight and easy to be integrated in more places (e.g. edge functions) 2. Repeated access for paid resources and subscription support (though not available atm) 3. Lifecycle hooks
🚀 After millions of payments on x402, we’re excited to introduce x402 V2. Listening to community feedback, we’re releasing V2 to evolve internet-native payments with greater flexibility and power. What’s new 🧵
0
1
4
Guess we are officially in a nostalgic, 90s-inspired design trend now that even x402 has started using Instrumental Serif 😂
🚀 After millions of payments on x402, we’re excited to introduce x402 V2. Listening to community feedback, we’re releasing V2 to evolve internet-native payments with greater flexibility and power. What’s new 🧵
1
1
2
One more thing you can stop managing by yourself
Introducing API Keys Public Beta Secure your routes with Clerk's managed API keys. Let your users create and revoke keys with prebuilt UI components.
0
1
4
Aftershocks keep coming
Researchers have found two new vulnerabilities in React Server Components while attempting to exploit the patches last week. These are new issues, separate from the critical CVE last week. The patch for React2Shell remains effective for the Remote Code Execution exploit.
0
0
0
bro is cooking nonstop 🔥
x402 Landing Page Agent creates on-brand landing pages for sale funnel or campaigns. just describe your campaign and get a live URL no design skills, vibe coding or hiring freelancers
1
1
2
🚨 Announcing LISA-Bench: Open Source Release @AnthropicAI proved AI can exploit smart contracts (55.88% success, $4.6M simulated). @AgentLISA_ai proved AI can defend them in production (95% success, $7.3M+ detected). The difference? LISA-Bench—#AgentLISA 's proprietary
8
16
47
If you're using React Server Components, please update immediately.
There is critical vulnerability in React Server Components disclosed as CVE-2025-55182 that impacts React 19 and frameworks that use it. A fix has been published in React versions 19.0.1, 19.1.2, and 19.2.1. We recommend upgrading immediately. https://t.co/kue7kd0XEX
6
28
437
I might have to repeat this: with @AgentLISA_ai over $7M in exploits have been detected on record https://t.co/Z88LaLjeYh , and that’s just since LISA’s public beta in late Jun 2025
agentlisa.ai
Every exploit below could have been caught with LISA's security analysis before deployment. In Web3, the cost of inadequate security isn't just reputation—it's everything. Secure your protocol with...
New on our Frontier Red Team blog: We tested whether AIs can exploit blockchain smart contracts. In simulated testing, AI agents found $4.6M in exploits. The research (with @MATSprogram and the Anthropic Fellows program) also developed a new benchmark:
0
2
4
Kudos to @AnthropicAI @MATSprogram for advancing benchmark standards. The field has long suffered from outdated datasets that miss real-world attack patterns. We're addressing this gap. @AgentLISA_ai uses a comprehensive benchmark built from actual security incidents and
ieeexplore.ieee.org
Smart contracts are decentralized applications built atop blockchains like Ethereum. Recent research has shown that large language models (LLMs) have potential in auditing smart contracts, but the...
New on our Frontier Red Team blog: We tested whether AIs can exploit blockchain smart contracts. In simulated testing, AI agents found $4.6M in exploits. The research (with @MATSprogram and the Anthropic Fellows program) also developed a new benchmark:
13
23
24
Two Extreme Misconceptions About Using AI/LLMs: The core issue is cognitive bias regarding engineering investment. The first group's problem is excessive optimism and overgeneralization. They might happen to get decent results on a particular case and immediately think they've
人们在使用AI/LLM时的两个极端误区，核心问题是对工程化投入的认知偏差。
0
3
14
We're releasing a visual agent & workflow builder ▪️ Fully open source ▪️ Built on https://t.co/tOVJiPK51X ▪️ Outputs "𝚞𝚜𝚎 𝚠𝚘𝚛𝚔𝚏𝚕𝚘𝚠" code ▪️ Supports AI "text to workflow" ▪️ Powered by @aisdk & AI Elements ▪️ Sample integrations (@resend, @linear, @slack) Clone &
100
141
2K
Built a x402 powered Nano Banana Pro Playground with @codenut_ai and @thirdweb
https://t.co/VfDu9asuXN You can do quite a lot with just 15 cents now. 😎
4
2
19
We let the Internet down today. Here’s our technical post mortem on what happened. On behalf of the entire @Cloudflare team, I’m sorry.
blog.cloudflare.com
Cloudflare suffered a service outage on November 18, 2025. The outage was triggered by a bug in generation logic for a Bot Management feature file causing many Cloudflare services to be affected.
553
1K
10K