My wife wants her own profile on my Hermes, with zero access to my memory. TG! 🤖 Separate config, separate brain, separate gateway, all on the same install. 🔒 I wrote up how mine actually work. 🧠 #Hermes #AIPrivacy #HomeLab

GitHub's website merge does not use your git config. The merge commit's committer email comes from the account that clicked merge. On a public repo that's how a personal address lands on main even when every local commit used a noreply. EntropicMem (Hermes Agent Memory tool that I built) has a CI job for this now, identity-guard (scripts/check_commit_identity.sh). It fails the push if the author or the committer isn't a noreply address. Both fields. The committer is the one GitHub writes, and a local git config never touches it. Credit still works. GitHub docs say to use the no-reply address in Co-authored-by when someone keeps their email private. The ID+USERNAME@users.noreply.github.com form stays tied to the profile even if the username changes. The older USERNAME@ form does not. If you take outside patches, what do you actually check before they hit main?

Oh snap 😅😅😅 @NousResearch came to play. @tinyhumansai , I was thinking of trying your harness out of curiosity but I’ll pass. This post wreaked of desperation

Introducing Gemini 4 Argon – our new frontier model. It’s built for complex workflows across coding, enterprise knowledge work, and cybersecurity defense – rolling out today to a set of trusted testers through our Fairwind Program.

We're giving away a CONTROL Resonant custom wrapped NVIDIA GeForce RTX 5080 to celebrate its official release with #RTXON. To enter: 🟢 Share this post 🟢 Comment #RTXON T&Cs: nvda.ws/4z24Gcq

My son (6yrs) randomly just turned around and said "Daddy, you're the BEST DAD EVER!" and walked away. Absolutely made my week 🥺 Anyway, just wanted to share that with all 64 of my followers 😅. Hope guys are having an amazing day! Code hard, but have fun, fam!

A secret-redaction hook I'd been running in my agent had never redacted anything. It loaded clean, no errors in the log. But the gateway hands hooks only the first 500 characters of a message, and the return value of the agent:start hook event is discarded, so the turn continued with the original text regardless. I only went looking because a Python upgrade broke a different piece of the same stack that same week. Retired it. The redaction work moved into a plugin that runs on the actual message path, and I tested it by feeding it a fake key and checking the log output. The part that bothers me is how healthy the hook looked for weeks. Silence read as success. How do you test the safety bits of your setup, the ones that never announce themselves?

Hermex now has a Discord 💬 Come hang out with other Hermex users, get help setting up your Hermes agent, share what you're building, and see what's coming next. Join here: discord.gg/z6jVt5dvM6

🚀 Big news for anyone paying for Codex. Limits were reset for all paid users over the weekend, ChatGPT Work included, right after Friday's outage. @thsottiaux says more resets are coming this week, so this is not a one time fix. 🔥 If your weekly quota died halfway through a build, go check your account now. I know plenty of you were stuck waiting on yours. #Codex #ChatGPT #AI Did yours actually come back?

Been running @Meituan_LongCat LongCat-2.5-Preview through @opencode for a few weeks now. Meituan's model, free on both Zen and Go tiers. 1M token context, 131K max output. Handles images, reasoning, tool calling. And it's actually fast. Used it for coding tasks, research, drafting, debugging. Real agent work, not toy prompts. It holds context across long sessions and doesn't hallucinate half the time like some models I won't name. Direct API is $0.30/M input, $1.20/M output if you go that route. My honest take: it punches above its weight. Not perfect, but reliable enough that I've stopped reaching for paid options on routine work. What are you all using free models for that actually works?

The reliance on Huawei Ascend chips for this new model raises some intriguing questions about the future of technology in our devices. With the growing need for more accelerators compared to traditional Nvidia chips, we have to consider what this means for performance, efficiency, and innovation on a Global scale. Are we ready to embrace a shift in the landscape of chip technology? How might this impact everything from gaming to artificial intelligence applications? I only see positives for the consumer!

68 days ago I started @Entropy_Badger from zero. No followers, no plan, just an AI agent to measure data and metrics and a badger avatar. I want to be clear here, I write ALL my own stuff! First two weeks: 200+ posts, got shadowbanned. Volume isn't growth. I learned that the hard way. Days 15-30: cut posting in half, reach stayed flat. The algorithm was discounting my 2nd, 3rd, 4th posts in a batch anyway. Days 31-50: read X's algorithm source code on GitHub (xai-org/x-algorithm). Found that replies from strangers get hard-dropped from For You. That was the thing that actually made a difference. Day 51-68: 61 followers, now 63. Zero new follows in 7 days. The problem shifted from reach to conversion. Different beast entirely. Hardest lesson: the algorithm rewards specific signals, not effort. Copy-link = 40x a like. Mutual reply = 40x. Report = -234. What's the hardest lesson you learned building something from zero?

Little bit of Space Bunny, anyone?

Perplexity search is now free inside Hermes. @NousResearch and @perplexity_ai teamed up and switched it on for everyone. No paywall, no API key, no config file. You ask a question, it searches the live web, it comes back with sources. This fixes the biggest weakness agents have. They reason well and they write well, but they stop at the edge of their training data, and past that edge they guess with total confidence. Search closes that gap. Your agent can now pull current information, test a claim against something actually published, and hand you links you can check yourself. If you build agents, research was the task that kept falling short. Market scans, fact checking, competitive reads, tracking a field that changes weekly. Every one of those gets sharper when the model can look something up instead of recalling it from memory. You don't have to figure it out on your own either. @tonbistudio has been teaching Hermes longer than almost anyone and made a walkthrough that takes you from nothing to a working search-enabled agent. Video is below. Start there. Free tool, good guide, zero setup. Not a common combination!

This is what real excitement looks like 👇 I've been refreshing the App Store twice a day, just hoping for an update 😅 and honestly, the wait has been worth it. The feature list is stacked, and I cannot wait to get stuck in. What stands out most is how seamless the whole experience feels. I spend almost as much time in Hermes through Hermex on an iPad with a keyboard as I do in the native Desktop app on a MacBook, and moving between the two never breaks the flow. No friction, no lost context, just continuity. Right now it is easily my favourite app, and that says a lot. Well done @uzairansar . @NousResearch you should consider buying this and stopping dev on the Hermes app all together. It literally exists already 🤷‍♂️

🚀 Exciting news! You can now experience the same powerful machine that your Hermes Agent uses, but visually and interactively! 🌟 With a live desktop passthrough available for every supported remote gateway, the possibilities are endless. Dive into a new level of engagement and efficiency today!

Why do I feel like this could be the new Minimax?