gonna update my OpenClaw🦞 bot

The rumours are getting crazier by the hour

The timing is crazy, literally the few things I wished Visual Studio Code had they’ve just shipped. - Unified chat experience bringing in Claude and Codex into their native chat system - Integrated browser for testing and debugging - Allowing Claude and Codex to use their native MCP

Claude: “You’re absolutely right!…” Codex: “Just say the word…”

Introducing Claude Opus 4.6. Our smartest model got an upgrade. Opus 4.6 plans more carefully, sustains agentic tasks for longer, operates reliably in massive codebases, and catches its own mistakes. It’s also our first Opus-class model with 1M token context in beta.

GPT-5.3-Codex is now available in Codex. You can just build things. openai.com/index/introduc…

Rumors are that Opus 4.6 was initially to be released as Sonnet 5. Interesting.

This looks really good 👀

4.5 to 4.6 doesn’t feel like much of an upgrade. Certainly great improvements but doesn’t feel drastic. 5.2 to 5.3 actually feels like a huge bump. It feels significantly more productive and far more intelligent than 5.2. The speed improvements are fantastic too.

Looking forward to testing some of these light weight OpenClaw systems.

Introducing QuickClaw ✨ I made the first iOS app that launches your own OpenClaw agent in under 30 seconds. No Telegram No API keys No setup One app, Sign in → your agent is live. In the appstore right now. Limited capacity, first come first serve.

The fact that Anthropic doesn’t follow the AGENTS.md or ./agents/skills standard is ridiculous. It’s so pathetic.

nearly all of the best engineers i know are switching from claude to codex

Claude 4.6 is like working with your best work friend Codex 5.3 is like working with the biggest autistic jerk... But he happens to be a fucking genius

I was loving Conductor until the force to use the new status view. Faaaaar too much noise. I work with one workspace at a time.

Do the little agent icon colours have any significance in Augment Code Intent? Trying to get my head around if they represent status, progress, agent type, tools etc?

I’m so glad companies are rethinking how we build. THE IDE was great, but we need something better now for the new way we build. This is a fantastic start.

I've been using Augment Code Intent the last 24 hours and it's pretty amazing just incredibly difficult to understand what's going on and when it's waiting on me or not etc. The agentic team/swarm orchestration system is fantastic, it works in Intent the way I would imagine it.

I really can't wait until I no longer have to pick thinking/reasoning levels. I just want the model to do this automatically. Just have a fast/lite model that makes this decision.

If Cursor allowed me to use my codex and Claude subscription ( and not via API key ) I would be in heaven.