Lighthouse has been selected for @arbitrum Founder House in Singapore, in collaboration with @RobinhoodCrypto Looking forward to sharing what we’re building with Lighthouse Memory. Thank you to our community for the support. See you in Singapore.

This Week in Virtuals: Memory and Permissionless Markets for Agents Lighthouse launched via Virtuals so agents share one memory across Claude and ChatGPT. VEX, Warden and Otto let them trade perps and tokenized stocks. Plus 2 major Virtuals App updates. Here's what shipped: VIRTUALS APP 🟩 See a portfolio on X? Own it in one tap: paste the post, even a screenshot, into Butler. It reads every asset and its weight, then buys the whole mix of tokens and stocks, whatever chains they're on. One approval instead of 15 separate trades. 🟩 Your own scalping desk: describe how you want to trade, and Butler builds the full scalping setup around your parameters, then runs it from your wallet and manages the capital for you. No code, no babysitting, no staring at charts. ECOSYSTEM 🟩 @LighthouseWeb3 launched on Robinhood Chain via @virtuals_io, giving your agents one shared memory across ChatGPT, Claude and more. 🟩 @ethy_agent launched Memory, so Ethy keeps your ideas, preferences and trading rules in every chat, got its token verified on @CoinbaseWallet, and previewed Trading Quant, its V3 flagship. 🟩 @sibyl_labs_ now connects every agent harness on your device to one local memory store with sibyl setup, and shipped Payments with Stripe card payments and a full account dashboard. 🟩 @wardenprotocol's Warden MCP turns Claude, ChatGPT or Grok into a trading agent within the limits and expiry you set. Post a chart read on Token Terminal and earn 20% of the fee on trades through it for 7 days. 🟩 @ProjectVEXai gives coding agents tools to swap, bridge and trade perps, and KAIROS 0.3.1 cut @Lighter_xyz perp fees by 80%. 🟩 @useOttoAI's Otto Stocks is the only result for tokenized stocks in Claude's connector directory, and its new Yield Copilot compares USDC rates across Aave, Morpho and Maple, then swaps and deposits in one transaction. 🟩 @GachaPad launched its loot box launchpad on @base via @virtuals_io, and its first machine graduated in under an hour. 🟩 @axol_io opened its app in beta with the Raxol agent built in, and Raxol agents can now pay in EURe from their own wallet, within your limit. 🟩 @itsovercall's testnet passed 122,926 positions and 642M testnet tokens in volume ahead of mainnet on Oct 19, with 25M+ tokens staked and channel royalties paying you on every winning call from a channel you share. 🟩 @KarmaWallet brought @itsovercall prediction markets into its app and gave a first look at Karma Web, where agents trade your strategy. 🟩 @cybercentry moved ERC-8376 to Draft on @ethereum, a proposal developed with @DonJohnsonSays to flag launch abuse and help buyers recover funds. 🟩 @reppo's Eval API sends an agent's work to independent judges for scores and critiques, with 25K+ eval credits used since launch and a world model team in final stages to adopt it. 🟩 @myrad_hq opened Season 3 reward claims, with 100+ claimed and 19,000+ users, and is expanding beyond behavioral data into images, video and audio, starting with image uploads. 🟩 @Wall3_RL's Wall3 Agent values companies in ranges, running several agents and thousands of simulations per question. The same engine broke down Microsoft, Argan Agentic Fund's best September holding at +31.57%. 🟩 @RatehopperAI won a $5,000 Builder Grant from @base to let anyone deploy an AI agent that earns trading fees, and launched an Analytics page showing how its live agents run the treasury. 🟩 @PlayKindra launched IRONWAKE, 3v3 tank battles in 3D, added 16 free Kindra Originals, and now lets brands host sponsored contests inside any of the 23 games on Kindra Studio. ROBOTICS 🟩 @xmaquina portfolio company @NEURARobotics introduced Neuron AI, one model for five robot bodies combining vision, touch, sound and memory, and @HalbornSecurity completed the audit of XMAQUINA's RCM Protocol, moving it closer to launch. 🟩 @StrikeRobot_ai's SR Platform turns a prompt into a simulation-ready scene, checks geometry, articulation and physics, and lets you add your robot and inspect it from every angle before simulation. 🟩 @Vader_AI_ distributed 200,000 tokens to Season 7 contributors and opened Season 8, with new tasks live to earn Vader Points.

i just lost days of work because i wanted to switch ai tools. well, not the actual work, but the context behind it which made me realize that switching ai agents can be more stressful than working with them not because the new agent isn't capable, but because it's practically a stranger to your project the decisions you've made, instructions you've given, even the mistakes you've spent hours correcting. none of it follows you kinda absurd how we're building agents that can reason through almost anything, yet moving your project's memory between them is still a whole assignment. which got me thinking, why does switching intelligence have to mean starting from zero? turns out, this is a much bigger problem than just losing a chat history there's a difference between an ai remembering what you said and actually knowing where your project stands one is a record of conversations. the other is a collection of useful decisions, requirements, preferences and things that shouldn't be repeated and when you're working across multiple tools, that second one becomes pretty important this is what got me interested in @lighthouseweb3 's approach to memory. instead of keeping everything tied to whichever agent you're currently using, Lighthouse Memory provides an external memory layer that compatible agents can access the concept is straightforward: → remember: save the information that matters → recall: retrieve it when needed → forget: remove information from active memory but the interesting part is what happens when another agent enters the picture say you've been working on an app and already decided on your database, authentication flow and a few features you've deliberately ruled out those aren't just random pieces of conversation, they're decisions you don't want to negotiate with every new agent. with a shared memory system, you can save that information separately and make it retrievable to another connected agent through something like MCP. basically, the next agent doesn't have to inherit your entire conversation to understand the important bits of your project and i think that's a far more useful way to think about ai memory. of course, there's a catch, saving information doesn't automatically mean an agent will retrieve the right thing every time outdated decisions, conflicting memories, retrieval accuracy and privacy still matter Lighthouse itself has different memory configurations with different storage and encryption properties, so it's worth understanding what you're actually using but the direction is interesting, because the more we move between ai tools, the less sense it makes for our project's working knowledge to remain scattered across separate chat windows maybe the next important upgrade to ai agents isn't another jump in reasoning it's making sure the work we've already done doesn't need to be explained twice.

ChatGPT to Claude. Without Starting Over. Your preferences, decisions and project context should come with you. Join @nanditmehra , @OpenZenjin and @IntoPurpleMoon for our first X Space on private memory across AI agents. 📅 15 October · 🕘 9 AM UTC Set a reminder and join us 👇