Mark Starr

@trq212 @hudaman

I'm going to be doing a skills livestream tomorrow, Thursday March 19th @ 11:30am PT to talk live about skills and how to use them. Come with questions! Special guest @hudaman from Uber to share how Uber is using skills.

@fubert_the_3rd @mattions_it @JJEnglert @openclaw Max

@MarkStarrSounds @mattions_it @JJEnglert @openclaw How do you avoid getting rate limited right away? Are you on Claude Max or pay as you go?

After a month of watching my fellow builders set up their @openclaw , I finally took the plunge this past week. Last night my agent ran overnight on a project we came up with together, and it was ready for review when I woke up this morning. It has its own GitHub account. Its own email. Its own Twitter. It runs 24/7 on an old MacBook Pro with the lid closed. And it has enough tools connected to actually do real work. But the magic moment wasn't the overnight build. It was something way simpler. I told it to message me at 7:30 AM with a daily plan. And it just did it. Figured out how to do it on its own. That "figure it out" mentality from an agent that actually has access to tools and a computer felt different than anything I've used before. For the first time, it felt like something capable of doing real stuff. Not a chatbot. Something else. And I'm just scratching the surface. It took me about 8 hours to get here. I want to help you get there faster. Here's everything I learned along the way, plus a prompt you can copy and paste into your OpenClaw once you're set up. Getting started I set it up on an old MacBook Pro. Dedicated device. You want this running independently so it does not have access to your data. Having a virtual device on @Hetzner_Online is also good. Installation took about an hour. Then I spent the next two hours having Codex tighten the security before training it anymore. Sandbox commands. Whitelist only what you need. Do this first. Then I hit a wall. It felt like a chatbot. Limited permissions. Couldn't access tools. Couldn't browse. It took another 2-4 hours to get terminal access and Playwright browser control working. I used Caffeinate in terminal to keep it running with the lid closed. I set up dedicated accounts. GitHub, email, Twitter. Give it its own identity so it can operate independently. Training it - Keep your Heartbeat.md lean. It gets read every session and burns tokens if it's bloated. Identity, active projects, key preferences. That's the hot cache. - Install a memory plugin early (ClawVault, Supermemory, or Lumen Notes). Persistent memory across sessions is what takes it from chatbot to something that knows your work. - Build skill files for recurring output. Emails, social posts, documents. Each gets its own file with format, voice rules, examples, and a checklist. It follows these like playbooks. - Define your agent's persona and tone. I built out voice files based on what I'd already created in Cowork and the output quality jumped immediately. - Point it at your existing repos. It can pull context from anything you give it access to. If you've already built structure somewhere, don't rebuild it. Reference it. Best advice I got from experienced OpenClaw builders Force plan before execution. Make it tell you what it's going to do before it does it. Saved me from multiple rabbit holes. Back up your repo to GitHub every night. Your config files, skills, and memory directory are the training. Lose them and you're starting over. Think in workflows, not one-off tasks. This compounds fast. I also applied the same repo structure from my Cowork setup guide: Your-Workspace/ ├── Heartbeat.md ├── Brain/ │ ├── about-me.md │ ├── brand-voice.md │ └── working-preferences.md ├── Skills/ ├── Projects/ └── Memory/ I'm about a week in. Still early. But I can see where this is going and I wish I'd started sooner. If you're just getting started, here's the prompt I'd paste in on day one to fast-track the whole setup: -- You are going to help me set up my workspace so that every future session starts with full context about who I am, what I do, and how I work. We're building the files and structure that make you useful from the first message. Interview me in phases. Ask questions, then build files based on my answers. Don't rush. Don't assume. Ask before you build. Phase 0: Foundation Check if I have a Heartbeat.md file. If not, create one. Keep it lean. Recommend a memory plugin for persistent context. Ask what tools I use daily and help me connect them. Recommend sandboxing and whitelisting commands from the start. Phase 1: Identity Interview me to create Brain/about-me.md. Ask about my work, background, what I'm building, and positioning. Show the file. Get approval before moving on. Phase 2: Voice Interview me about how I want my agent to sound. Phrases I use. Phrases I'd never use. Tone shifts by context. Create Brain/brand-voice.md. Get approval. Phase 3: Working Preferences What I want help with. Communication style. Workflow pain points. Output preferences. Create Brain/working-preferences.md. Get approval. Phase 4: Skill Files For each type of recurring output, create a skill file in its own folder under Skills/. Each gets: format, voice rules, examples, quality checklist. Ask what I create most often before building. Phase 5: Active Projects Current projects, goals, deadlines. Individual files in Projects/. Phase 6: Memory System Update Heartbeat.md with a summary of everything we built. Create Memory/ directory with subfolders for people, projects, context. Add glossary.md. Phase 7: Reference Sources Any existing repos, docs, or files I want referenced. Organize access. Rules: One phase at a time. Show each file before saving. If unsure, ask. Concise files. Lowercase, hyphens, .md format. Start with Phase 0.

@mattions_it @JJEnglert @openclaw /reset works or /new worst case. If it’s stuck and you’re stumped spin up CC CLI to fix it or have your bot fix itself. 9/10 times this works. Opus 4.6 fwiw

@JJEnglert @openclaw May I ask which model you are using to drive the agent? In my experience I started with codex, then moved to openrouter free, but then I started to have serious problem of quality output. I was not impressed. In particular, when it gets stuck, how do you unstuck him?

@ErikSchluntz Very cool!

While most of my time is management now, shipping this was a fun side quest :) Now to stay in flow while claude is working you can ask 1-off side questions without interrupting it with /btw!

@stash_pomichter @HisamSabouni @dimensionalos Love moondream. Crazy stuff.

@HisamSabouni @dimensionalos we use moondream for this exactly! local and fast

Openclaw can now understand physical space and temporality. Integrate with any lidar, stereo, rgb camera. Fully open source. Video below is our openclaw on a Unitree G1 humanoid. We integrate with most drones, quadrupeds as well.

@nappytheplug My god we are entering another E-8305 era aren’t we 😩

every touring DJ on earth is having nightmares about their sticks not reading, we are watching several people experience this pain every weekend all these tech nerds in music, and there’s no solution for this? who remembers back in 2011 when the CDJs would just *turn off* 😆

@tomieinlove That’s not trolling — it’s corporate masking with impact! 🚀

Tried to speak like ChatGPT to annoy people but now I’m getting promoted and networking better and opening new doors in my professional life

@GavinNewsom @Dodgers Considered giving you my vote but you’ve finally crossed the line with this one.

What a game! Congrats, @Dodgers!!! You make California proud!

@SFGiantsMemer Ohtani becomes the first player in MLB history to hit a home run off of himself in the 9th to win it - opening a rift into the Dark Beyond and dragging us into a spiritual battle between angels and demons that WILL determine your bloodline. This is the level of glaze I expect.

What are your Game 7 realistic predictions? Mine: Dodgers win 8-2 with Kershaw getting the final out while Joe Davis cries and is consoled by John Smoltz.

@SFGiantsMemer Saw someone else mention this and it got me thinking 🤔

Remember when they used to intentionally walk power hitters to keep them from hitting home runs?

Here’s why I don’t like music. It’s overstimulating almost immediately to me. It makes me physically dizzy and disoriented when I hear most music. Secondly, most people who are super into music are into it the same way annoying Marvel fans without a personality are into Marvel.

@GiantHotTakes Please. I beg. We need that RF expertise again.

Mike Yastrzemski is a FREE AGENT Would you be interested in a reunion?

@AndrewCurran_ So bothersome. Generally makes me feel uncomfortable when I spot things like this. Same experience as hearing a note out of key 😬

The strangest thing about seeing model outputs everywhere is wondering how many people even notice. Maybe ten percent? Language is not just a carrier of meaning, it is a scaffold for many things. This is a live experiment, so we will discover the second order effects together.

@ConwayAnderson @NTWR_LaL @moondreamai Makes sense. Most of what I have seen has been PPE and collision detection, which is seems simpler by comparison. My field is EHS so I am pretty excited to see the ways this sort of thing can improve accident prevention, especially in mfg settings. Neat times.

@NTWR_LaL @MarkStarrSounds @moondreamai RL can help but generally harder the more ambiguous something is if there’s some definitive “unsafe” vs “safe” thing to detect it’s more possible

Blazingly fast visual reasoning from an open model that runs everywhere. Moondream 3.

@NTWR_LaL @moondreamai There’s value in identifying injuries quickly in a mfg setting. Seconds matter.

@moondreamai I think it is not targeting actual use cases. I don’t want to know where is injured person(that event already occurred). I want the intelligence for possible injury from the scene understanding.

@JWonggg 🤔

If MVC2 came back on new gen platforms you buying it right? TBH I probably would buy 100 copies minimum to give away

@nicebloom42 @Bomb_xyz @NoContextHumans You only get charged once, not each way. Still insane.

@Bomb_xyz @NoContextHumans Yup. $8 one-way

The Oakland-San Francisco Bay Bridge averages 260,000 vehicles daily, each paying a $8 toll.👀💰

@GiantHotTakes At this point bring in therapists for the team. Something ain’t right.

Just a lifeless effort. Nobody wants to be there. Depressing

@matthewk36711 Still time to delete this

Wilmer Flores has had an illustrious career with the #SFGiants, and fans adore him. But he's showing some immense signs of regression, and once Jerar Encarnación finishes his rehab assignment, he should take the DH role. Let's take a look into Flores' decline: A thread 🧵

@PavlovicNBCS Wilmer this, Lee that - barely anyone seriously discussing Yaz. Why?

The Giants lost 5-3. They left San Francisco in first place and they'll fly back home in third after a two-game sweep. The lineup has been carried by two or three guys all month, and it's catching up to them.