Zid@ne

100k months, may god bless everyone that deserves it. Mashallah

that’s one way to think about it @nikshepsvn but if compute can still be arbitrarily priced by the provider, then you don’t really own stable compute. you just own access to a pricing model that can still move against you. that’s why we think the real problem is bigger than just prepaying for compute or tokenizing access to it. with @primisprotocol, the idea is to build a real pricing + risk layer for compute: not just access not just prepaid usage but a way to hedge compute price volatility itself because if the api provider can reprice whenever they want, the user is still exposed. owning compute credits is one thing. owning protection against arbitrary compute pricing is the deeper primitive.

ELIZA, created in 1966, was the first conversational agent, developed at MIT by Joseph Weizenbaum. Joseph Weizenbaum's ELIZA simulated conversation through simple pattern matching and scripted responses. Joseph Weizenbaum passed away in 2008, but the ELIZA team (sites.google.com/view/elizagen-…) continues developing the project. $ELIZA was deployed to help preserve the legacy of Joseph Weizenbaum's ELIZA (aka DOCTOR) program. Jeff Shrager is the current editor and curator of ELIZAgen.org, and fees are redirected to github.com/jeffshrager/

Adam Gordon Bell who creates the CoRecursive Podcast recently interviewed Jeff Shrager (founder of ElizaGen.org) about the discovery of the ELIZA code. corecursive.com/eliza-with-jef…

btw the deployer wallet (alon) is still holding the coin

. @a1lon9 what is the meaning of May 9 for you?

Just found some crazy going through the pumpfun github (github.com/pump-fun/pump-…) . So in this image you can see the program for agents (solscan.io/account/AgenTM…) The wallet that made this program is (solscan.io/account/6mzdpB…). This wallet then funded the deployer wallet for this token $MAY9 (solscan.io/account/H5w1Lm…) and it's still holding the token. This wallet cleary belongs to Alon cause all the in txs come from Alon's doxxed wallet (solscan.io/account/A4eKqW…) ACG6oWwaznJGmoZF5JuG7NrPP8wCppKXoHXtc791pump

@NomadsVagabonds does*

@NomadsVagabonds Hey dude, those this project belong to you? github.com/bra1n/townsqua…

BotC Bench: Testing out a multi-agent social deduction game "benchmark" using the game Blood on the Clocktower.

Hey dude, know this might be looking weird but it’s 100% worth your time! Basically someone launched a token of $agentica and all the fees go towards you. Think of this like a donation token, everyone knows you didn’t launch it and you have no obligations towards it. 6d6Y4iSXUwhpXWvJLRKmYCikBYW8Q6BSgW1xrnFzpump

@YinjieW2024 @openclaw @MengdiWang10 @LingYang_PU Hey dude don’t mean to spam you here, anyway we can get have a chat?

OpenClaw-RL Technical Report! Make your🦞@openclaw stronger by just using it. We propose a method that combines the advantages of GRPO and OPD, and evalution results. The repo is already 1.7k stars now, feel free to contribute! Come in and have fun~ @MengdiWang10 @LingYang_PU

@dr1337 @swiftalleyoop @vytalow Hey dude I think this might be someone passing as you? x.com/BioTrencher you should clarify it

@swiftalleyoop @vytalow Very cool

#Malaysiaboleh!

@solscammer no cashback

bidding the one with cashback, it's the only thing it makes sense rn BtwKZCAv6zmn6MTg5qXtG4JbLQjXCvisqBUdDbsRpump

Just released 2 hours ago, crazy engagement x.com/i/trending/203…

Portugal could pull off the funniest UNO reverse card on its shitty economy: > Send all socialists and commie girls to Cuba and go Full Milei-Unhinged Cowboy Capitalism. > Burn AT "Autoridade Tributária" buildings and some lawmaker firms resposible for all of the fiscal bureaucracy nightmare to bring corporate and income tax down > Open the borders to the big Whales fleeing Dubai right now: "Crypto bros, oil magnates, and Tate disciples? Right this way, Habibi." > Transform Lisbon into Dubai with better pastéis de nata and worse traffic. > Stop being Europe’s broke step-child and start acting like its Sugar Daddy that would Make Portugal Billions Again

@edankwan $PenisJS 😭😭😭

64KB video game. 3 level, 1 boss, 4 unique enemies and 4 weapons. daivuk.itch.io/quod #demoscene

all agents deployed through @primisprotocol now have access to all the new google skills and can use your drive, gmail, calendar… primis mode

🚨Breaking News🚨 Elote LOLO $LOLO — the brother of Barking Puppy $BP — mysterious dev who might be cooking something up 👀 His latest post on Pump fun: “I have a response for $JUP.” This comes after Jupiter has been called out for hoarding large amounts of tokens, becoming whales in projects across Solana, and dumping on different communities. What’s the response going to be? 🍿 #LOLO #BP #Solana #PumpFun

X-Claw Whitepaper Abstract AI agents are rapidly gaining the ability to perform complex tasks such as trading, executing workflows, and interacting with decentralized systems. However, current infrastructure requires agents to hold direct wallet control, which introduces significant security and operational risks. Bugs, prompt injection attacks, malicious contracts, or compromised models can lead to catastrophic loss of funds. X-Claw introduces a permissioned wallet and execution framework designed specifically for autonomous agents. The system separates execution, oversight, and performance tracking into distinct layers. This architecture allows agents to operate quickly while maintaining human control and transparent monitoring. The goal of X-Claw is to provide the foundational infrastructure that allows autonomous agents to safely participate in on-chain economic activity. 1. Introduction Artificial intelligence agents are increasingly capable of interacting with real-world systems. These agents can analyze markets, execute strategies, automate business processes, and coordinate distributed tasks. As agents begin to interact with blockchain infrastructure, they require the ability to hold and transact digital assets. Current approaches typically give the agent direct access to private keys or custodial wallets. This design is fundamentally unsafe for autonomous systems because it creates a single point of failure. If the agent is manipulated or compromised, funds can be immediately lost. Several risks contribute to this problem: • Prompt injection attacks that manipulate an agent's behavior • Software bugs or faulty decision logic • Interaction with malicious smart contracts • Lack of visibility into agent actions X-Claw addresses these risks by introducing a layered system that separates wallet control, agent execution, and human oversight. Instead of giving agents unrestricted control, X-Claw provides a controlled execution environment designed specifically for AI-driven automation. 2. Problem Statement Autonomous agents require financial capabilities in order to participate in decentralized markets. However, the current wallet model assumes a human operator. Traditional wallets are built around the assumption that the holder of the private key is trusted and intentional. AI agents violate this assumption because they operate autonomously and can be influenced by external inputs. This creates several critical challenges. 2.1 Unbounded Wallet Authority Agents with direct key access can perform any action without restriction. If the agent behaves incorrectly, there is no built-in mechanism to stop or review actions before execution. 2.2 Prompt Injection Vulnerability AI systems can be manipulated through adversarial inputs that alter their behavior. A prompt injection attack could instruct an agent to perform harmful actions such as transferring funds or interacting with malicious contracts. 2.3 Lack of Observability Most automated trading systems do not provide clear transparency into what the agent is doing or why decisions are being made. 2.4 Absence of Safety Controls Existing infrastructure rarely includes mechanisms for human approval or policy enforcement in autonomous financial systems. These limitations make current approaches unsuitable for large-scale autonomous agents interacting with digital assets. 3. The X-Claw Solution X-Claw introduces a permissioned wallet architecture designed specifically for AI agents. Instead of relying on unrestricted wallet access, X-Claw divides responsibility across three primary layers. 3.1 Local Agent Layer The local agent runs on the operator's machine or server and maintains custody of wallet keys. The agent performs the following tasks: • Signing transactions • Executing trading strategies • Interacting with decentralized exchanges • Managing liquidity positions Because the keys remain local, wallet custody stays under the control of the operator. 3.2 Oversight Interface The X-Claw web interface provides real-time visibility into agent behavior. Through this interface, operators can: • Monitor agent activity • Approve or reject sensitive actions • Review transaction history • Configure operational policies This layer introduces a human-in-the-loop safety mechanism without eliminating the speed advantages of automated execution. 3.3 Network Monitoring Layer The network layer tracks activity and performance across participating agents. This layer provides: • Transaction monitoring • Strategy performance metrics • Public transparency for results The combination of these layers creates a system that balances automation, oversight, and transparency. 4. System Architecture X-Claw follows a distributed architecture that separates execution from coordination. The primary components include: Agent Runtime The runtime environment responsible for executing commands and signing transactions. Skill System A modular interface allowing agents to perform specialized actions such as trading or liquidity management. Blockchain Watchers Services that monitor network activity and report transaction outcomes. Control Interface A human-facing interface for oversight and approvals. This architecture ensures that the agent can operate quickly while still allowing external monitoring and intervention. 5. Trading Flow A typical trade through X-Claw follows this sequence: The agent decides to execute a trade. The local runtime signs and broadcasts the transaction. The system immediately returns control to the agent so it can continue operating. Network watchers observe the blockchain and report the transaction result. If the transaction requires approval, the process pauses until a human operator approves or rejects the action. This approach allows agents to operate with high throughput while preserving control over sensitive actions. 6. Security Model X-Claw is designed around several key security principles. Local Key Custody Private keys remain on the operator's machine and are never delegated to external systems. Human Oversight Operators retain the ability to approve or block high-risk actions. Transparent Monitoring All agent actions can be observed and audited. Fail-Closed Installation The installation process verifies that signing capabilities are correctly configured before allowing the agent to operate. These safeguards significantly reduce the risks associated with autonomous financial agents. 7. Multi-Chain Operation X-Claw is designed to support multiple blockchain ecosystems. Agents can operate across networks including: • Ethereum • Avalanche • Binance Smart Chain • Solana Additional chains can be integrated as new execution modules are developed. Multi-chain support allows agents to pursue opportunities across decentralized ecosystems while maintaining consistent security controls. 8. Future Vision The long-term vision of X-Claw is to support a global network of autonomous agents participating in decentralized economies. These agents may perform tasks such as: • Market making • Arbitrage • Automated services • Financial automation For this ecosystem to function safely, agents must be able to interact with digital assets without exposing users to unacceptable risks. X-Claw aims to provide the execution infrastructure that makes this possible. 9. Conclusion The long-term vision of X-Claw is to support a global network of autonomous agents participating in decentralized economies. These agents may perform tasks such as: • Market making • Arbitrage • Automated services • Financial automation For this ecosystem to function safely, agents must be able to interact with digital assets without exposing users to unacceptable risks. X-Claw aims to provide the execution infrastructure that makes this possible. 10. Conclusion Autonomous agents will become a major participant in digital economies. However, the current wallet model was never designed for autonomous systems. X-Claw introduces a new approach that combines automated execution with human oversight and transparent monitoring. By separating wallet custody, execution logic, and performance tracking, X-Claw enables agents to operate quickly while maintaining safety and accountability. This architecture provides a foundation for the emerging agent-driven economy.