If you have hardware like this that’s a very different conversation; my message is directed to the 95% case that can’t spend 5k or 100k on hardware.
@steipete
-
Gateway Security Model: One User Multiple Agents Isolation
By
–
What would you use this for, why is isolation needed? The security model is 1 gateway = one user = 1…n agents
-
Evaluating AI Claims: Separating Hype From Real Use Cases
By
–
Try it for yourself and make up your mind. There’s defo some claims that are inflated, lots of real good use cases too!
-
PR merge skills added to agent scripts repository
By
–
Ah just checked, the PR merge skills are in there now. Not sure why this was done, some of these are in my agent-scripts repo though as well.
-
OpenShell and NemoClaw: Building Secure AI Agents with NVIDIA
By
–
Been so much fun cooking OpenShell and NemoClaw with the @NVIDIAAI folks! Huge step towards secure agents you can trust. What’s your OpenClaw strategy?
-
Blocklists and Platform Governance: Unintended Consequences
By
–
no, too opinionated and I might get things wrong on some. These blocklists killed other platforms
-
OpenClaw and acpx Codex App Server Released for Coding
By
–
OpenClaw + acpx/codex app server is for coding. Just watch tomorrow’s release!
-
Anthropic Enhanced with Adaptive Thinking and Million Token Context
By
–
Not at all, we added adaptive thinking and 1Mio context support to support Anthropic even better!
-
Plugin-Based Architecture Encourages Community Contributions
By
–
We worked very hard to make everything plugin-based so absolutely, send a PR!