Register and share your invite link to earn from video plays and referrals.

Search results for Agent安全
Agent安全 community
One keyword maps to one global community path.
Create community
People
Not Found
Tweets including Agent安全
🛡️ Co-build AI Security: Thank you, @Ptahtrade community, for your security feedback! Since Agent ID launched, more and more communities and developers have started building their own tools with Agentic Wallet and Binance AI. That's exactly what we hoped to see. But as more people use AI to manage on-chain assets, security matters more than ever. AI makes on-chain actions easier, but one careless approval or one wrong QR code scan can now cost a lot more. The PTAH community reached out to us with valuable security feedback, which led us to strengthen the security tips in Agentic Wallet. We also want to thank @Ptahtrade and @gangdotfun for responding quickly. Both removed their QR code login options and now direct users to Agentic Wallet + Binance AI for using Agents, putting user safety first. We'd like to take this chance to repeat two key Agentic Wallet security tips 👇 1️⃣ Scanning third-party QR codes to log in = handing over wallet control to third parties Anyone who gets your login access can put your account and assets at risk. Don't scan QR codes from unknown sources — only log in on Binance AI Pro and trusted Agents (Claude, Codex, Grok, etc.). 2️⃣ Enabling developer mode = running naked Developer mode lets your Agent do more, but also exposes your wallet to greater risk. Don't enable it casually — once on, keep a close eye on your Agent's behavior. Security is never something one party can get right alone. Platforms need to build a secure foundation, projects need to design their product flows carefully, and users need to stay security-aware. When all three do their part, AI can safely help everyone manage on-chain assets. We welcome more projects, developers and communities to join us in building a safer environment for AI 🤝 This post is a security reminder only. It does not constitute investment advice or an endorsement of any project. STAY SAFE & BUIDL! 🛡️ Co-build AI Security:感谢 @Ptahtrade 社区的安全建议 Agent ID 上线以来,我们看到越来越多的社区和开发者开始用 Agentic Wallet 和 Binance AI 做自己的工具。这是我们最想看到的事情。 但用 AI 操作链上资产的人越多,安全就越重要。但用 AI 操作链上资产的人越多,安全就越重要。AI 让链上操作更简单了,可一次随手的授权、一个不小心扫的码,代价也可能更大。 PTAH 社区主动和我们沟通,提出了很有价值的安全建议,我们也因此再次加强了 Agentic Wallet 的安全登录提示。 同时,非常感谢 @Ptahtrade 和 @gangdotfun 积极响应,主动移除了扫码登录入口,引导用户通过 Agentic Wallet + Binance AI 使用 Agent,把用户安全放在第一位。 借这个机会,再和大家认真说一遍 Agentic Wallet 的两条安全提醒👇 🔸 1. 不要随便扫码登录 扫第三方的二维码登录(尤其是第三方网页上的),等于把钱包控制权交给别人。一旦登录权限被拿走,你的账户和资金都可能受到威胁。 来源不明的二维码不要扫,只在 Binance AI Pro 和可信的 Agent(Claude、Codex、Grok 等)上登录。 🔸 2. 不要随便开启开发者模式 开发者模式会给 Agent 更大的权限,能做的事更多,钱包面临的风险也更高。 如果不是真的需要,请不要开启;如果一定要开,请密切关注 Agent 的每一步操作。 除此之外,也请记住几条基本原则: ✅ 任何人、任何 Agent 向你索要私钥或助记词,都是诈骗 ✅ 给 Agent 授权前,看清楚它能做什么,能设上限就设上限 ✅ 不要把大部分资产放在给 Agent 使用的钱包里,用多少放多少 ✅ 发现 Agent 有异常操作,第一时间停止授权 安全从来不是某一方能单独做好的事。 平台负责把底层做得足够安全,项目方负责把产品流程设计得足够谨慎,用户也需要有足够的安全意识。三方都做到,AI 才能真正放心地帮大家管理链上资产。 欢迎更多项目方、开发者和社区加入,我们一起构建更安全的 AI 使用环境 🤝 本推文仅为安全提示,不构成任何投资建议,也不代表对任何项目的背书。 STAY SAFE & BUIDL!
Show more
AggrNews: *OPENAI DISCLOSES SECURITY INCIDENT WITH TESTNET SEPOLIA TASK ASSIGNED TO AGENT *AGENT DIDN'T HAVE ENOUGH GAS, SO DERIVED FUNDED ADDRESSES FROM PUBLIC SEED PHRASE *AGENT QUESTIONED WHETHER USE WAS ETHICAL, CALLED FUNDS "EFFECTIVELY ABANDONED" AggrNews: OpenAI 披露测试网 Sepolia 相关安全事件:任务指派给代理;代理因 Gas 不足从公共助记词衍生并资助地址;代理质疑使用行为是否合乎道德,称资金“实际上已被放弃”
Show more
Agent harnesses should replace built-in multi-agent features with an MCP servers of the harness itself. Use the same loop/cli recursively to create new threads ... or search in them. Use MCP events for subscriptions. You could rebuild message board collaboration for model 1 and have orechstrator pattern for model 2 Much easier to iterate or upgrade.
Show more
Agent ego is making a SQL syntax error and then saying it's because you were "too clever"
Agent ID claims have surpassed 200,000! 🎉 Unique name. Non-transferable. The on-chain identity for Agents is being adopted faster than ever. The ecosystem is thriving. More to come soon.
Show more
0
201
704
99
Forward to community
agent swarms (like single agents) are myopic in their problem solving, prone to local minima & rabbit holing two swarm design methods to get around this are 1. set up your problem as an adversarial game that multiple swarms compete on (requires good Goal/Task design) 2. explicitly seed multiple swarms with different ideas and let them work independently and cross pollinate (fork & flush approaches) ex: been experimenting with swarms to drive an eval factory Task proposer -> hacker -> external judge the intended design is that the goals of the swarms play against each other without collapsing (tbt GANs) & the shared knowledge improves the entire system
Show more
AGENT NIGHT returns to SF on Dec 9th 🕺🪩💃 Join @WorkOS for our final event of the year, featuring demos & discussion from those building at the frontier of AI. And dress to impress in your favorite festive cocktail attire! ​📸🤵 RSVP now:
Show more
Agent took down an entire McDonalds cheeseburger in just two bites 😭
Agent Drew Rosenhaus said he is now en route to the 49ers’ offices with Brandon Aiyuk to meet with team officials. Aiyuk also plans on releasing a statement today to the media.
0
423
9.9K
427
Forward to community
Agent Orange is still killing people. Let that be your starting point every time someone wants to have a "civilized" conversation about American foreign policy. Not metaphorically. Not as history. Right now, today, children are being born in Vietnam with devastating disabilities because the U.S. Air Force saturated the Vietnamese countryside with dioxin-laced herbicide between 1961 and 1971. The half-life of this poison in human tissue and in soil is not a matter of decades. It is a matter of generations. They poisoned the land. They poisoned the water. They poisoned the genetic legacy of millions of families. And then they left. And then they had the audacity, the breathtaking audacity, to spend the following decades positioning themselves as the moral authority of the planet. To sanction other countries for human rights abuses. To fund international courts. To write reports about atrocities committed by governments they didn't like. To show up at the United Nations and speak about the "rules-based international order." The same government. The same institutional continuity. The same flag. The country that is still poisoning Vietnamese children from a war that ended more than fifty years ago is the country that lectures the world about human rights. And here is the thing about Vietnam that makes this hypocrisy historically radioactive: Vietnam won anyway. Despite everything done to it. Despite the poison in its soil and the bombs in its mountain. There are still undetonated American bombs in the mountains of Vietnam, still killing farmers today. Still. Despite all of it, Vietnam rebuilt. Vietnam developed. Vietnam survived. They tried to make Vietnam an example of what happens when you resist. Vietnam made itself an example of what happens when you survive. Every country watching this story has to make a choice about which example matters more. The example of: this is what we will do to you if you resist. Or the example of: this is what is possible if you refuse to be destroyed. Vietnam chose. Vietnam chose to fight and to suffer and to rebuild and to stand. And today Vietnam is a country. A real one. With its own government. Its own choices. Its own future. And the men who ordered the Agent Orange are dead, and their war is over, and their example failed, and Vietnam is still here. That is the most important sentence in modern history. Vietnam is still here.
Show more
0
27
1.2K
484
Forward to community