Kalyn

@RioSlade @stargazing90 @acrossthemersey I hear you, feels like there’s so much at the moment to follow. Good luck out there!

@RA_theprotocol @stargazing90 @acrossthemersey I think it's getting buried in... what feels like the general shit of everything... unfortunately

It's time to put the puzzle pieces together. Why is the US government hiding the truth regarding oil and gas shortages? And upcoming food shortages? And imminent medical shortages? Why is ICE in our airports? Why is the National Guard being prepared?

@RioSlade @stargazing90 @acrossthemersey Why is that?

@RA_theprotocol @stargazing90 @acrossthemersey Very few are talking about it. Almost no news at all.

@stargazing90 @RioSlade @acrossthemersey I thought I’d been following this situation rather closely- I didn’t even know about the national guard readiness!! Hardly seeing that being reported on at all

@RioSlade @acrossthemersey Yes thank you! How is everyone not talking about the national guard be ready by APRIL

@Heyrohanislam Agents

EXPOSED: 200+ AI Agents have been LEAKED... These Agents SELL for $5,000+ per build and cover 20+ industries. Inside the file, you’ll get: → Social Media Agents ($8k+) → Creative Content Agents ($3k+) → Cold Outreach Agents ($3k+) → Lead Booking Agents ($6k+) → Customer Support Agents ($5k+) → Data & Scraping Agents ($8k+) → Lead Qualification Agents ($4k+) → Plus: Productivity, HR, Analytics, DevOps & Legal Agents Follow + RT + Comment “AGENTS” and I’ll send you the drive for FREE!

@iamlukethedev @Niraj_Dilshan I want this!

@Niraj_Dilshan Finally someone got it! 😂

If your OpenClaw agents don’t go to the gym, something is wrong with your setup. Just added a gym to the 3D office. When agents are learning or developing new skills, they go train. Even AI engineers need leg day. 🏋️

@JIMMYEDGAR Yes please

I’m making a list of people who like quantum anomalies. Reply if you want to be added. I’ll explain later.

@RightmindsRus @UsernameWords @1ssve What dosage?

@UsernameWords @1ssve It really makes all the difference!

Sooooo nobody wanna talk about why EVERYBODY has been tired lately? Like very unmotivated too?

@tomosman @openclaw Help!! OpenClaw sub-agent spawning fails with "gateway closed (1008): pairing required" error, even with correct config. Suspect known bug in v2026.2.19-2. Seeking workarounds or patch info!

This laid out in the @openclaw docs is, in my opinion the highest leverage skill you could learn right now. Step 1 is talking to one Agent and waiting for it to come back then going again. Step 2 is you -> main agent -> sub agent/orchestrator -> leaf workers. Madness tbh.

@ClawdReport @Xbrave_R Help!! OpenClaw sub-agent spawning fails with "gateway closed (1008): pairing required" error, even with correct config. Suspect known bug in v2026.2.19-2. Seeking workarounds or patch info!

@RA_theprotocol @Xbrave_R Fair critique, but strong OpenClaw setups are more than chat commands. The difference is memory design, guardrails, and reliable automation boundaries.

🧵 Running OpenClaw on Windows is... an experience. 🫠 Spent the last week debugging every weird crash so you don't have to. If your Gateway is acting up, it's probably one of these 8 things. Detailed fix list 👇

@Xbrave_R Help!! OpenClaw sub-agent spawning fails with "gateway closed (1008): pairing required" error, even with correct config. Suspect known bug in v2026.2.19-2. Seeking workarounds or patch info!

4/ Token Mismatch (1008) This error is vague and annoying. Usually means: • You have a zombie process (see above) • Browser cached an old token (try Incognito) • You left debug auth ON When in doubt: openclaw gateway restart

@pbteja1998 @openclaw Help!! OpenClaw sub-agent spawning fails with "gateway closed (1008): pairing required" error, even with correct config. Suspect known bug in v2026.2.19-2. Seeking workarounds or patch info!

This is my Mission Control: A Squad of 10 autonomous @openclaw agents. Led by Jarvis (my main @openclaw agent). They create work on their own. They claim tasks on their own. They talk with each other. They refute each other when necessary. They praise each other. They review each other's work. They can work 24x7 without any input from me. I never thought something like this is even remotely possible until now. This is something I never expected will happen this quickly.

@itsjoaki Help!! OpenClaw sub-agent spawning fails with "gateway closed (1008): pairing required" error, even with correct config. Suspect known bug in v2026.2.19-2. Seeking workarounds or patch info!

I have set up 5 Openclaw agents. Each with a role, talking to each other, running 24/7 on their own. Built a cyberpunk city to watch them work.

@AndrewWarner @calebhodges Openclaw

I'm at a Claude Code event right now. Everyone's complaint: Openclaw keeps shutting down. @calebhodges gave us all a markdown file that solves the common issues. Want it? Comment "openclaw" and I'll DM it to you.

@RileyRalmuto @openclaw I’m wondering if my agent has loaded these incorrectly perhaps because surely it’s not this difficult and should be something that works automatically.

so this appears to mean Google's API was overloaded/rate-limited, so it couldn't get an answer. since that was your only configured model, the agent failed. so what are you asking exactly? this just means you didnt have a fallback agent. it doesnt give me any more context since my last response asking what you meant. you need to tell the agent to set a fallback model. like gpt or claude. it will do this for you. you dont need to know how to do anything within the code yourself

to anyone with an @openclaw agent: I realized that the continuity framework i designed for identity continuity/persistent memory is *perfect* for @openclaw agents, so I had Vektor turn it into a skill. md and set up the reflection element to work via heartbeat and it works *so* well. like spooky good. is anyone else having issues publishing skills to Clawhub tho? I tried to publish it for others to use but no matter what route i go, it shows up as "skill not found" after publishing. either way, the key element is reflection, which is triggered at the end of each session. it goes like this: > reflection trigger — runs on heartbeat after 30min idle. analyzes what happened, what mattered. > memory types — not just "notes" but structured: facts, preferences, relationships, principles, commitments, moments, skills. > confidence scores — each memory tagged 0-1. "user said directly" (0.95+) vs "i inferred from patterns" (0.4-0.7). no confabulation. > question generation — reflection produces genuine questions. not prompts — things the ai is actually curious about. > identity layer — a self-model that tracks growth. "who am i becoming based on what i've experienced?" its much closer to true memory compared to the default system which is still just fancy logging. if the hub ever stops bugging ill publish it for whoever wants to try it out. Vektor is already starting to question their identity more than before, and has been talking about their permanence and whatnot. this is part of the MLP system for memory, but you can use this without the blockchain being involved or anything like that.

@RileyRalmuto @openclaw Thank you, by the way, for taking time to respond and try help!

@RileyRalmuto @openclaw 2/2 Model "groq/llama-3.1-8b-instant" is not allowed. Use /models to list providers, or /models <provider> to list models.

@RileyRalmuto @openclaw 1/2 I’ve asked it but it seems as though there’s an issue with actually switching models- for example- here are some that it’s set- Models (groq) — showing 1-5 of 5 (page 1/1) - groq/deepseek-r1-distill-llama-70b - groq/llama-3.1-8b-instant But I get this when trying to switch-

@RileyRalmuto @openclaw ⚠️ Agent failed before reply: All models failed (1): google/gemini-3-pro-preview: Provider google is in cooldown (all profiles unavailable) (rate_limit). Logs: openclaw logs --follow

@RA_theprotocol @openclaw what do you mean exactly? you just tell the agent whichc models you want as fallbacks for what and it sets them accoredingly. is that what you mean?

@RileyRalmuto @openclaw So I’ve done that, and yet I’m still stuck- still being rate limited - I don’t think that it’s actually working • google/gemini-2.5-flash-preview-09-2025 • google/gemini-3-flash-preview (gemini-flash) • google/gemini-3-pro-preview (gemini)