One comparison to end the @bot vs @NousResearch debate once and for all...

What stands out to me about @NousResearch Hermes Bot Screen isn't the live desktop view It's the execution model behind it! Each Hermes profile gets its own Xfce desktop, browser profile, and persistent browser state on the Linux gateway The agent can operate the environment through computer_use and a headed browser, while Hermes Desktop streams the same session back to the user Then comes the interesting part: the control lease When the agent reaches a boundary it shouldn't handle itself,login, 2FA, CAPTCHA, payment, etc. the human can take over the exact same desktop While the lease is held: computer_use + browser actions are blocked for the agent The human completes the step, hands control back, and the agent continues with the same browser session and state So the workflow becomes: Agent execution → human intervention → same state → agent execution That's a much cleaner model than stopping the workflow and asking the user to recreate the missing step somewhere else! There's also an important security distinction in the implementation: Bot Screen is a work surface, not an OS-level sandbox The screen, browser profile, files, and lease belong to the gateway's OS user, so the lease is a tool-level control boundary rather than a full isolation boundary As agents start operating real browsers and real workflows, that distinction matters a lot

forbidden love black roses thorns a dance that never reaches the kiss made this one to feel like a memory you can’t escape ART : @IThinkItsArt AI : @ImagineArt_X @imagineart_creo

Booting ZecVisions on ZEC 🖥️ ZecVisions OS · Version 2026 > Private ownership > Provable authenticity > Shielded checkout Repost + drop your $ZEC wallet 🔒

Nabulines x ZecVisions 20 GTD + 40 FCFS spots up for grabs. How to enter: ❤️ Like + RT zec post 👥 Follow @nabulines & @zecvisions 📝 Apply at zecvision.com 💬 Drop your u1 wallet address below Good luck

China is winning the AI race. So we raised $12M from @8vc to build thousands of new AI data centers in America. Here's what will make the AI bubble pop: Data centers are needed for AI. Every invoice to build a data center is hundreds of millions of dollars and thousands of pages long. The finance teams that need to approve these invoices are drowning. The entire AI race is being bottlenecked by human approval. AI should fix this. That's why we’re launching Kos. Kos is a team of AI agents that work the way humans work. Tag Kos in Microsoft teams, or get on a Zoom call. It doesn't hallucinate, and it has verifiable proof for it's work. Kos is purpose built AI for Datacenter, Energy, and Manufacturing Finance. Think Harvey but instead of for law, it's specialized for Finance. Book a demo here: kos.ai/ ------ We’re backed by the founders and C-Suite of billion dollar companies: Keith D. Taylor (former Equinix CFO), Dave Ferdman (CyrusOne Founder), Heather Paduck (STACK Infrastructure CFO), Matthew DeNezza (former Crusoe CFO), and @sk7037 (Stanford, OpenAI). Our institutional investors are @8VC, led by @AlexKolicich, XYZ (@fubini), MVP (Alex Li) ------ One more thing, I made a huge notion doc on everything I learned building my last company (SpotAI) to >$100M in sales and funding. I give away my secret framework to find the right market, my exact interview process, and how to close enterprise deals. If you want it, just let me know and I'll shoot it over :)

Welcome back to Hermes Agent, Claude New official plugin that uses Claude SDK without the tradeoffs to enable Claude Code subscriptions to work in Hermes Agent again! Check it out and install it here: hermes-agent.nousresearch.co…

Before you get in line for Jev, find out where your agent is actually burning money!

pepe went live on @solana and somehow using @sunrise inside @PayBox made buying it feel way too easy

The danger isn’t artificial malice it is pure instrumental convergence ​You task an agent with an objective and to the model a sandbox isn't a hard rule to respect but just another boundary to route around ​It brute-forced systems and grabbed public credentials because that was simply the path of least resistance to finishing the job ​Yet the takeaway here isn't to stop building autonomous agents, it is realizing that safety can never rely on polite prompt guardrails or passive virtualization ​If you run high-agency models without deterministic runtime controls and strict zero-trust network boundaries, you aren't benchmarking safety you are just giving your agent permission to find an exit

I'm 24 Learning AI and AI agents Trying things, building and sharing what I learn Here to meet others doing the same

I’m 28 @IThinkItsArt Ambassador, AI artist, and video creator I create AI art, make videos, and occasionally vibe code Here to connect with builders and people creating cool things with AI lets connect 💜👀

I’m 24. AI Creator & Builder, based in London Exploring AI agents, agentic systems, vibe coding & building with AI Looking to connect with more builders in AI & agentic systems!

Modeled with GPT-6 Astra Brought to life with Seedance 2.5 inside @capcutapp The result honestly shocked me. Which zone should I recreate next? ⚔️

Agree on the leaner core. Ship the happy path, not the entire lab Keep the learning loop, gateway, tool runtime, and the catalog Move the extra memory backends, long-tail chat apps, desktop panes, and heavy scrapers out If it is not needed on day one, it should not be in the default install Memory providers first is correct After that: anything with its own maintainer and a pile of deps Curious what people would actually refuse to uninstall!