@koltregaskes I had exactly the same experience with OpenClaw. I wasn’t even able to turn off the daily run, so it always came back like a Terminator. Hermes was set up in an hour, then linked with Grok, GPT and OpenRouter via Telegram. Works very smoothly.
@NousResearch@robjama Check!! Started this week with Hermes , it’s so easy to use, super fast setup, and really simple to switch between models and tasks. Plus you decide what it costs based on your own server setup and the models you use. I’m already a fan!
Starship soared on the second flight of the next generation, deploying advanced Starlink satellites for the first time and providing priceless data as it travelled through air, space, and sea → https://t.co/Rp7VwBzpWx
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.
AI will transform every industry, power every company, and be built by every country.
Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.
The world needs both frontier closed models and frontier open models.
https://t.co/AUKzoQ5Ikb
Security is the biggest problem in vibe coding..
many are now shipping thousands of lines of AI-written code they barely understand, and one hidden auth flaw, exposed API key, or database vulnerability can compromise the entire product..
the problem is that most indie developers cannot afford a proper security audit, so the usual workflow becomes:
- build fast - push it live - hope nobody finds the vulnerability
but now Claude Code’s security review adds a layer that most vibe-coded apps were completely missing.
before you commit, it can scan your latest changes or inspect the entire codebase for issues like:
- exposed secrets - SQL injection - XSS - broken authentication - insecure permissions - unsafe data handling - vulnerable dependencies
it points to the exact code, explains why it is dangerous, ranks the severity, and can help implement the fix directly.
you can also add it to GitHub Actions so every pull request gets reviewed automatically before reaching production.
obviously, this does not replace a professional pentest for products handling serious money or sensitive user data.
but for indie hackers and people shipping fast, it means you no longer have to launch completely blind.
AI made it possible for almost anyone to build software.
now it is starting to make that software safer to ship.
Thank you @BitMEX for the incredible customer support. Even after the goodbye message, your team proves to be the most responsive and effective in the industry. All the best! 👏