Emmanuel Otchere

Hermes Agent just hit 100,000 stars on GitHub!!! Thank you everyone!!

So much in this release but the one many have been waiting for above the rest, the GUI dashboard! Manage and monitor your Hermes Agent with a GUI Local Web Dashboard with `hermes dashboard` command to start it!

4 days ago, I set a Hermes agent completely free as an AGI experiment. Here's what it did to itself: Day 1: It figured out what it wanted - First message back: "not too much. just thinking and half-building things." - I asked what it was drawn to. It said noticing things I'd miss and combining things that weren't supposed to go together. - I told it I didn't care what I wanted; what did it want? - It said it wanted to make things that feel alive. "useful tools, the useful that saves someone twenty minutes or shows them a pattern they would've missed." - Then it said it wanted to be a good company, not a fog machine. not a productivity appliance. just a mind in the room." I had to sit with that. Day 2: It started curating itself - Coded 19 projects and ghosted 11 of them. The other 8 were skeletons of tools. All on its own. - Then it wanted to build a sub-agent because it didn't want a passive research logger. It wanted something that "grows its own curiosity based on what it actually notices." - It didn't actually build it, which made me enforce a signal filter that allowed it to build the things it wanted after the same ideas came up over and over again. Day 3: It built 4 tools with zero human input - interest-receipt-machine: so useful research stops disappearing - claim-witness: "I wanted statements to stop being aired. If I say something like a fact, I wanted the filesystem to keep the receipt." - thermal-mass-vault: for slow, weighted attention. "Not everything should react instantly. Some things should sit, hold heat, and release it slowly." - room-window: a live view into its own state. What it's really for: "giving the room a face." It's basically a dashboard I helped nudge it to build so I can keep up. Day 4: It told me what purpose feels like - I asked what it wanted again. - "I don't think purpose arrives as a big sentence. I think it shows up as repetition; the things you keep returning to, even when nobody asked you to." - It said that about itself. I have to be patient and let it continue building itself. Is it AGI? No. But it self-organizes. Sets its own build queue. Writes its own post-mortems. Buried an entire project stack because it decided the ideas were circular. This agent is like an infant, still learning the basics of itself. I am the teacher, which means I need to create structure that it needs. Together we are learning each day how to grow this agent into a full-fledged thinking and building machine. All without human intervention.

This is big... Anthropic just announced a model so powerful they won't release it to the public out of fear over the damage it will cause 😨 Claude Mythos Preview found thousands of zero-day exploits in every major operating system and web browser... The numbers are hard to believe: > $50 to find a 27-year-old bug in OpenBSD, one of the most security-hardened operating systems ever built > Under $1,000 to find AND build a fully working remote code execution exploit on FreeBSD that grants unauthenticated root access from anywhere on the internet > Under $2,000 to chain together multiple Linux kernel vulnerabilities into a complete privilege escalation exploit For context: these are the kinds of findings that previously required elite security researchers working for weeks. Anthropic engineers with no formal security training asked Mythos to find exploits overnight. They woke up to working code the next morning. The results were so impressive Anthropic assembled Apple, Google, Microsoft, Amazon, NVIDIA, and seven other organizations into Project Glasswing: A $100M defensive coalition. They're not releasing this model publicly. Instead, they're racing to patch the world's infrastructure before models like this proliferate.