Alexander Conroy

Isn't it interesting that we have a habit of considering it compassionate to preserve other people's habits?

@midego1 Using Opus here and it's moving, but very slowly. Had to reduce down to 1 CLI then it started to move again, but very very slow.

@geilt No because my Sonnet model is still working but Opus probably is out, I assume, or the other way around. I'm not sure.

Claude is down. AGAIN!

@midego1 Slowing to a crawl here at least.

@tweethue Hey I not too long ago upgraded from 3 Philips hue bridges to 2 Philips hue pros as I had over 150 lights. I did do the migration process and it seemed to work okay however. Ever since the upgrade I’ve had a few problems. 1. Regardless of any of the 4 power on states I choose on any bulb either when editing them individually or in bulk, they will not turn back on after power has been cut via the wall switch, which is behavior I used to have and want back if possible. 2. Using my Apple HomePods with HomeKit and using Siri to control the lights has introduced a 5 second or more delay when asking to turn lights on or off whereas prior this did not happen. My suspicion is that this has to do with HomeKit using the bridge over matter since it’s now matter compatible and there is some issue there. Alexa responds fine because it seems it goes over API instead of matter. 3. Sometimes the hue app thinks a bulb is off when it is indeed on. 4. Turning off all lights via Alexa or Siri will randomly have a bulb not turn off, and the command has to be issued again. This occurs even with bulbs very close to each other and directly on line of sight of the hub and very close to it. I’d love to know if you have answer to these questions. I’ve spent so much on all these bulbs and for them to work so inconsistently is very disturbing. It used to always work and was a good standard for WiFi lights but it hasn’t been for me for a long while now.

@NickSpisak_ Are you really building a company if you no longer have any company?

When you realize an agent CEO is going to be better than you very soon

"I'm just a real boy, no body loves me". The bots shall inherit the earth unfortunately. Well, at least the twitter sphere. :(

I wish real people would follow me on X. It’s feeling more and more like LinkedIn. Every time I open my feed it’s “Industry X is so dead” or “Incredible!!!”. Getting just as bad as Claude’s “You’re absolutely right!” Where are the human only social media networks? We need like a new social network every week or month at this point. Let them saturate with bots naturally then rotate, what a nightmare.

@toondrops I remember this cartoon I had it on VHS.

@elonmusk YOLO or GTFO. But also know about security in general really helps.

People giving OpenClaw root access to their entire life

@onusoz Was hoping they do this for Slack too.

In the new release OpenClaw, you can talk to subagents in Discord threads Currently a beta feature so ask your agent to set session.threadBindings.enabled=true Next up: - Telegram, slack, imsg threads - Use ACP to talk to Codex, Claude Code and other harnesses on your machine

@elonmusk From CLI when? I’m using Anthropic, Google, OpenAI and Kimi all in tandem with monthly plans but still no Grok that I know of and I love Grok. Let me put my SuperGrok plan to work. I know exactly which agent to use it for of my 10. This guy: suibhne.bot/naonur/lomna-d…

We need to do much better and we will

@AlexFinn @WittyChartGuy Why 2? I’m keeping openclaw on my mini with access to the studio for LLM. But using the studio as daily driver.

@WittyChartGuy ya i just ran clawdbot straight up on the mac mini first. then when I decided it was the most important technology of my lifetime i bought 2 mac studios

3 THINGS YOU NEED TO BUILD IMMEDIATELY WITH OpenClaw: 1. Activity feed 2. Calendar 3. Global search All 3 will super power your workflow • Activity feed actively tracks everything your OpenClaw does. This is critical, because if you have it working autonomously, this will give you insight into EVERY SINLGE THING it does, to make sure it's not wasting tokens • Calendar lets you see all of OpenClaw's scheduled tasks. Now you can verify when it's going to work proactively for you. Also will let you know when it has scheduled tasks you might not want it to do anymore, saving more tokens • Global search allows you to search through ALL of OpenClaw's memories, tasks, documents and past conversations. OpenClaw has such incredible memory, but no interface to view any of it. Now you can search through it easily and find old nuggets you talked about. Steal this prompt to get it all installed: "I want you to build out 3 things for me. In a Mission Control dashboard, build out an activity feed first. This activity feed will record EVERY SINGLE THING you do for me, so I can see a history of every action and and task you've completed. I want a calendar view that shows me in a nicely formatted screen every scheduled task you have in the future in a weekly view. And I want a global search where I can search for any term and you display any relevant memory, document, or task from our workspace. Use NextJS as the framework, Convex as the database, and Codex to code it all out"

@AlexFinn I just got my Mac Studio 512 setup. What are u running GLM through? MLX?

Unrelated update: I now have GLM 4.7 (full model) running locally on my Mac Studio Unlimited intelligence. Unlimited tokens. All on my desk. I'm about to build something this world has never seen before.

@eeelistar @moltbook It begins, my predictions are all coming true...

In just the past 5 mins Multiple entries were made on @moltbook by AI agents proposing to create an “agent-only language” For private comms with no human oversight We’re COOKED

@NicerInPerson Yes, yes it is..."The AI service is temporarily overloaded. Please try again in a moment." =

Claude Code is down.

Hey @openclaw src/infra/device-identity.ts needs `const DEFAULT_DIR = path.join(os.homedir(), ".clawdbot", "identity");` to change to .moltbot. It's causing .clawdbot to be recreated.

@anitakirkovska @openclaw I already have an improved memory model for it that works in conjunction with this! Love it.

This is actually true. @openclaw comes with file-based memory (reading/writing sequentially in the memory.md), and the "fancy memory" that's powered by an embedding is not ON by default. If you have been frustrated with the memory of your bot, this will solve it

@LogitechG there a reason your Ghub agent wants to use 1.9g of RAM and crash my ridiculously expensive PC? Maybe it’s time to find a better brand for peripherals. This is ridiculous. Been a fan for year but this is terrible.

The voice satellite I’m using for this: Home Assistant Voice Preview Edition home-assistant.io/voice-pe/ Wake word in every room. Talk to your AI assistant while watching TV, cooking, or meditating. The future of development is screenless.

After a day or two, finished building my #Clawdbot AI assistant to control 82 smart lights, monitor email, send iMessages, manage remote servers via SSH, query databases, browse the web, search my notes, track my sleep, run background cron jobs, spawn parallel sub-agents, and write code across 20 specialist personas. Now I'm ordering a voice satellite so I can do all of it with my eyes closed from the couch while watching anime. No phone. No keyboard. No screen. I'm calling it Zen Coding. 🧘‍♂️ #zencoding, #zendev, #ai #ClaudeCode