SPFAdvisors

@sama Bro, I see a lot of hate come your way, let me just say thank you for the amazing tech you and your team are building, id say this is the most impactful thing in my lifetime to be created, besides my wife and kids lol but next , def next lol.

@steipete Brother, I see so much hate on you lately, let me be the one to tell you I appreciate you and what you've contributed to the world, its truly amazing tech and inspirational the ups and downs make us all better!

@steipete @keenon Your^ 😅

@steipete @keenon Even with the kinks appreciate you work bro, the tech is amazing! Anyone not seeing that is outtta touch with reality, this truly made Ai intriguing again, it was getting stale for a while.

You reply to everybody but now you’re afk huh? I updated to 4.29 by clicking “update” button in web ui and broke everything . Why? @steipete why would you do this?

@Voxyz_ai I’ve been fighting since 26 it’s responding but limping along, memory has slipped into dementia it comes and goes, cron jobs not firing, one of my agents says his heartbeat is now empty 🙄 it’s been a rough couple days

this version isn't working for me right now. symptoms: telegram shows typing, then pauses, then the answer arrives late. reproduces in other entry points too. it still looks like the queue + tool prep + prompt assembly stage that's stuck. a separate per-turn agent-prep / main-thread blocking problem remains. core-plugin-tools, model-resolution, auth, stream-setup are the suspect areas. model itself only runs 6.7s, queue + prep eats 44s. trace from one message: → 08:06:22 message.queued (telegram typing fires immediately) → 08:06:39 session lane starts processing, waited 16.8s → 08:06:59 agent stream-ready, prep 12s (core-plugin-tools 6s + system-prompt 2s + stream-setup 2.3s) → 08:07:06 run_completed, model itself only runs 6.7s → 08:07:11 message.processed, total duration 51s added docker data to #73428 too. anyone else hitting this? github.com/openclaw/openc…

@steipete this last update to 2026.4.26 has caused a lot of breaks and errors, I get this message frequently after most request anything that takes over 3 min to complete "No additional reply from me here."

@shipper_now Shipper

We raised $0 to kill AI slop. Introducing Shipper: the world's first AI agent with taste. RT+ comment “Shipper” and we’ll randomly send free credits.

@claudeai @AnthropicAI you get violations now for trying to set up a website and a Facebook page ? API Error: Claude Code is unable to respond to this request, which appears to violate our Usage Policy (anthropic.com/legal/aup). Try rephrasing the request or attempting a different approach. If you are seeing this refusal repeatedly, try running /model claude-sonnet-4-20250514 to switch models. this makes no sense at all

@HeyGen AvatarV

We solved character consistency. Forever Avatar V captures you in 15 seconds and holds your identity across every video. Change the look, outfit, and setting to create unlimited versions of you. RT + comment "AvatarV" below and I'll DM 100 credits to test it out (must follow)

@Delta service in general is getting worse every flight, as a skymiles memeber for 14yrs and Diamond Medallion who only flys first class it’s disheartening to watch and be a part of, the boarding process is struggling as well.

@steipete 😂

This guy emailed me asking for a *token session refund* because his claw made mistakes. 🙃

@elonmusk any chance we can try out grok with openclaw ?

@Strongpointfin1 using OAI is fine for openclaw

@steipete so im just hearing open is allowing us to use our pro sub to power open claw is this true? I really dont want to get my GPT banned ?

@dashboardlim gemini

gemini 3 pro is google's smartest model n8n is the easiest automation platform together, they can make up workflows that cost $10k-20k the integration lets you: → build AI agents with little or without code → automate image and document processing → create complex workflows in minutes → configure thinking levels and API settings i made a playbook that covers everything on: - gemini 3 pro + n8n setup - building decision-making agents - advanced configs nobody talks about most people don't know this exists yet learn it before they do comment "GEMINI" and i'll send you the playbook for building AI workflows with gemini 3 pro and n8n

@kat_kampf @GoogleAIStudio Access please

We started internal testing some big updates to the @GoogleAIStudio experience today! Coming to you early next year but reply below if you’d like early access in the coming weeks 👀

@SenWarren Where was this criticism under your control?

This holiday season, families face a trifecta of economic pressures: Record-high grocery bills. The weakest seasonal job market in over a decade. Rising electricity costs. Working people are struggling in Trump's economy, but he's too busy building himself a ballroom to notice.

@BernieSanders How do you want to combat so many things but are agains weapons of combat 🤔😴🤫

In an authoritarian society, a handful of oligarchs own media. Here’s where we are in America: Musk owns X. Zuckerberg owns Facebook & Instagram. The Ellisons own CBS — and are eyeing CNN & TikTok. Bezos owns Twitch & the Washington Post. We must combat oligarchy.

Want our full prompt guide? Retweet this thread and we'll DM it to you 🎥