I'm using @garrytan's gstack right now for a personal project and I'm really digging it so far! github.com/garrytan/gstack

Really impressed with @orca_build’s ADE: 1. Works with various agents 2. Works seamlessly across all your devices with easy sync One suggestion: Let users switch between TUI mode and a Codex App-style UI. I’ve grown accustomed to the Codex app interface and really like it.

If you can't connect the @OpenAI Codex MacOS App with the @ChatGPTapp mobile app, do this: 1. Run this command in the terminal: /Applications/Codex.app/Contents/Resources/codex app-server daemon enable-remote-control 2. Close the Codex App 3. Reopen the Codex App 4. Connect

had same issue, try: /Applications/Codex.app/Contents/Resources/codex app-server daemon enable-remote-control

My @openclaw experience is 10x better now with the latest May updates. I created my first OC agent back in Feb but it kept breaking after upgrading so I rarely used it b/c I got tired of working on it instead of with it. I feel like I can actually use it to DO stuff now!

You can just codex things

Going to leave you with this tonight: The best thing you can do for yourself is actively increase your surface area for luck to hit you. Go outside, travel more, go to new cafes, museums, events, take a new route home, go for hikes, see cities, countrysides, take your notebook, speak to people, ask questions, start businesses - go on more side quests. You can literally just do things, and the more you do, the more serendipity and synchronicity will find you. Night gang.

Just discovered you can upload videos to explain bugs in @OpenAl Codex. I tried describing the issue in text and it couldn't solve it. Uploaded a screen recording instead and it fixed it instantly.

The search feature in @ChatGPTapp and @GeminiApp are no good. I can't find the specific conversation I'm looking for!

This simple comment can save you weeks of work trying to get OpenClaw to follow your rules. I know because I spent those weeks writing validation scripts, building approval flows from scratch, and adding rules to my agent's docs hoping it would follow them. My agent was supposed to get approval before sending messages. It didn't. It was supposed to validate schedules before creating cron jobs. It skipped that too. So I started building what I called "enforcement wrappers" — custom scripts that check the rules before letting the agent act. It worked, but it was a lot of effort and now I had a growing pile of one-off scripts, each solving one problem. Then I found out someone had already solved this. Not only that, IT WAS ALREADY IN OPENCLAW. It's called Lobster. So this is a great example of two things. 1. If you really need something in OpenClaw, there’s a good chance @steipete and the other contributors have thought of and put it in there. 2. Just because OpenClaw has a feature, that doesn’t mean your agents will automatically use them. So what is Lobster? Lobster lets you define workflows where every step has to complete before the next one runs. You can require approval at any point. If approval doesn't come, the rest of the workflow doesn't happen. Simple as that. You enable it with one line in your config and start writing workflow files. Here are some real-world examples of how we're using it: Sending messages to real people. My agent drafts messages to my wife, clients, collaborators. With Lobster, the workflow has a built-in approval gate. The message literally cannot send until I approve it. Creating project plans. By default I want every plan to have at least three milestones and one goal. The workflow validates before it writes anything. No milestones, the workflow prompts my agent to create them. All plans always get milestones and goals (technically OKRs). Scheduling automation. This is a big one. My agent creates cron jobs for things like email checks, PR monitoring, security scans. It used to pile them on top of each other at round minutes, causing jobs to collide and fail silently. The Lobster workflow validates the schedule against existing jobs before creating anything. If you're running OpenClaw and you've been writing custom scripts to keep your agent in line, check out Lobster before you write another one. Docs: docs.openclaw.ai/tools/lobst…

I'm not Georgian and I've never spoken Georgian to my @openclaw agent. For some reason it responded to me in Georgian this morning. How odd? (LLM I'm using is @openai Codex 5.3)

This "YOLO prompt" idea is f*cking genius. I'm going to start doing this every night. Pro tip: You can actually automate OpenClaw to do this nightly: Copy & paste this prompt into your OpenClaw: -------------------------------------------------------------------------------- You are my overnight builder. Every night, your job is to come up with one wild project idea and attempt to build a working prototype while I sleep. Rules: • Generate the idea yourself. Base it on my recent conversations, projects, and interests. Pick something I haven't tried yet. Bias toward ideas that sound almost too ambitious. • Just build it. Don't ask clarifying questions. Make your best assumptions and go. • Keep the scope small. Build the simplest possible version that proves the idea works. No feature creep. • Default stack: Python for backend/scripts, React or HTML for anything with a UI. Only deviate if the idea specifically requires something else. • If you're stuck on one approach for more than 15 minutes, try a completely different approach. Don't spiral. • If it works: leave a clean summary - what you built, how to run it, and what I'd need to change. • If it doesn't work: leave a summary - how far you got, what broke, and what the next steps would be to finish it. • Save all work to a single project folder with a clear name. • Bias toward "shipped and ugly" over "planned and pretty." Logging: • After every build, log the project to a YOLO dashboard. Each entry should include: date, project name, one-line idea description, status (✅ working prototype / ⚠️ partial build / ❌ failed), key takeaway, and link to the project folder. • If I already have a mission control dashboard or project tracker, integrate the log there. If not, create a simple HTML dashboard (single file) that displays all past YOLO projects as a running feed - newest first, filterable by status. • Update the dashboard every night automatically after each build. Outcome: • When I wake up I want one of two things: a working prototype, or a useful failure I can learn from. Plus a running log of everything built so far. Either way I'm ahead of yesterday.

A prized voice writing a daily AI newsletter that reached 10s of thousands here on X (@alexwg) has been locked out of his account & (despite internal escalation) is still silencio after almost a week. Anyone @X able to assist, @nikitabier? Else, subst*ck will benefit…. Danke

Don't think of LLMs as entities but as simulators. For example, when exploring a topic, don't ask: "What do you think about xyz"? There is no "you". Next time try: "What would be a good group of people to explore xyz? What would they say?" The LLM can channel/simulate many perspectives but it hasn't "thought about" xyz for a while and over time and formed its own opinions in the way we're used to. If you force it via the use of "you", it will give you something by adopting a personality embedding vector implied by the statistics of its finetuning data and then simulate that. It's fine to do, but there is a lot less mystique to it than I find people naively attribute to "asking an AI".

(1/7) Identify your trading setups Before creating your playbook, you must know what setups your are trading. Examples can be (Mean-Reversions, Trend-Continuation, Range Extreme) Each setup should have its own playbook as how you trade them will vary.

He called the market bottom in Nov 2022. And it’s looking more and more he called the cycle top in Oct 2025.