David Lutz
467 posts


Thanks to @steipete and the @Alibaba_Qwen team! Love having local qwen3-coder:30b as my fallback model to codex! I think my @openclaw now has multiple personalities - surreal!

English

@aclfestival You got 1 good band. Are you just going to tweet this until October?!? Do better!
English

@sama Models have become really good. But need realtime info. Asked GPT5.5 about best text to image models and the answer was 9 months out of date
English

Wow. I just asked my @OpenClaw what we’ve actually built, and the answer was… a lot: ~207k lines of authored code across apps, skills, automations, dashboards, Android work, and little tools that now quietly run most of my life. 🦞
English

@thefox @antigravity Love this. Excited to see it improve and roll out to more tickers (like SHOP)
English

Just shipped AI-powered key moments in Google Finance so you can:
+Understand major price swings in stock charts (1 month+ view) 📈
+Jump to key moments in corporate earnings calls 🎧
Small team built these features from the ground up (and fast) using @Antigravity, check em out
finance.google.com/beta
English

Thanks @steipete ! It's inspiring to know that you recognize the challenges and are not just going to keep creating fixes one paper cut at a time. OpenClaw has become so valuable!! openclaw.ai/blog/openclaw-… via @openclaw
English

@m91michel @openclaw I think there's another regression in v .29 -- I'm also seeing some (not all) messages exist in ControlUI, but not in Discord
English

These last few @openclaw updates have been rough for Discord users. Broken slash commands and now with .27, no replies at all. To fix the latter check: messages.groupChat.visibleReplies and set it to: "automatic"
English

@lennysan Recently replaced Loom with recordly.dev. Opensource and keeps getting better! Love the automatic zoom feature! @webadderall
English

A few lesser-known products I love and use a ton:
1. cleanshot.com for screenshots
2. supercut.ai replaced Loom for me, so clean
3. textexpander.com for frequent snippets
4. matthewpalmer.net/rocket/ for emojis
5. brain.fm for focus
English

@hubermanlab I just want to say THANK you for this amazing podcast! I stumbled across it a month ago, and it's now my #1 listen. You are doing a great service to humanity! 🙏
hubermanlab.com
English

@adamsilverman @perplexity_ai I made it my default from the day it launched. Though, in the past 2 months, I've found that best results are often selecting Gemini Pro as the model. Lately, I've found myself using Gemini in Chrome more and more. And OpenClaw for browser automation
English

Anyone else using @perplexity_ai comet as their daily browser?
I haven’t been able to find anything as powerful.
It is by far the best browser agent on the market.
English

@LutzDave We already did some tweaks for that and inject more !
English

@steipete Do you have any suggested updates to OpenClaw bootstrap files with the new GPT 5.5 prompting guide? (and yes, i did ask openclaw/codex first - but suggestions looked too invasive)
developers.openai.com/api/docs/guide…
English

Love the "working..." with lives updates in @openclaw on Discord! Kudos! ❤️🦞
English

Image generation got less fussy: gpt-image-2 works through Codex OAuth, OpenRouter image models work through image_generate, and agents can pass quality/format/background hints when supported. docs.openclaw.ai/tools/image-ge…
English

OpenClaw 2026.4.23 🦞
🧠 GPT-5.5 lands
🎨 Image gen/editing: Codex OAuth + OpenRouter
🧵 Forked-context subagents
💬 Telegram, Slack + WhatsApp polish
🛠️ Install/update fixes
Sharper models, smoother claws.
github.com/openclaw/openc…
English





