David Butler

@samwoods Umbrella term for autism, ADHD, etc

@davogones what is neurodivergent?

I want more higher quality connections If you're: - into philosophy of mind/tech - doing legit AI and Agent work (no hype psychos please) - AI agent builders - Builders/founders - love writing essays (with or without use of LLMs) - High agency person - tryna run your 7-9 figure business as best you can If any of these sound like you, say hi below

@blueprint_os @Glomadic_lyfe @heavensbvnny The agents live on the same file system so I just have them peek into each other’s workspaces if they need to share context.

@davogones @Glomadic_lyfe @heavensbvnny What's your setup for coordinating between those three? We run 5 agents off shared memory files but curious how others handle it.

ADHDer's dream job: Coming up with brilliant ideas and immediately handing them to someone else to execute.

Built a little web based harness on top of Codex App Server. Was able to implement HTML-based chat. Excited about the possibilities.

@Glomadic_lyfe @heavensbvnny I use OpenClaw, Claude Code, and Codex. I build AI agents for neurodivergent people: divergentsystems.ai

@davogones @heavensbvnny which agent/framework do you use?

@TheGodfessional @heavensbvnny cal.com/david-butler

@davogones @heavensbvnny David please hook me up

@enjojoyy I build OpenClaw agents for non technical people and charge $77/month for hosting. divergentsystems.ai

What's the easiest way to host an autonomous agent for a non-technical person currently? (Openclaw, Hermes, Pi etc.)

@E__Strobel @petergyang Thanks for sharing this. What does your rollback plan look like?

Yes. Updating is the one thing I do myself. 1) Use both Codex CLI & OpenClaw doctor to do a pre-upgrade health check 2) openclaw backup create --verify 3) openclaw update --dry-run 4) openclaw update --yes --no-restart After the update, 5) openclaw config validate 6) again run both openclaw doctor and Codex CLI to check health 7) restart the gateway & check status I have literally never had OpenClaw break after updating when I do this. (I *have* had Codex CLI find & fix issues which likely would’ve broke my ‘Claw.) Yes, it’s more work in the short term, but it gives me better visibility on the upgrade process.

Is there a better way to update OpenClaw than do this? Shit breaks half the time when I tell my bot to update itself.

@jordan_ross_8F CLAW

The agency owners who installed OpenClaw two months ago are already adding millions in extra revenue and saving hundreds of hours a week. The problem is most agency owners don't have time to figure out the install, the security risks, where to start, or what to actually hand it first. So my team built a 111-page playbook that does it for you. Inside: — 93 copy-paste automations across ops, content, lead gen, sales, delivery, and hiring — The exact prompts to hand your agent on day one — 11 security fixes most agencies skip (and pay for later) — The hosting setup, the model routing, the partner system — Plus the April 2026 stack: Claude Managed Agents, AI Visibility tracking, the GEO playbook Your competitors are sleeping on this. Comment CLAW and I'll send it.

@helloiamleonie Working on a substack post about it. Will post here when it’s live.

@davogones That’s impressive. Any tips for skills, MCP or specific setup?

Agent harness for writing instead of coding. Anyone working on that? Any recommendations? (Please don’t tell me Claude Code unless you have a workflow that’s actually good)

@eliana_jordan I build OpenClaw agents for people. divergentsystems.ai Happy to help. OpenClaw has matured but remains really hard to run and upgrade. I’m mainly using Claude Code or Codex locally right now.

installing openclaw today finally am i too late?

@nbaschez DMed

Absolutely incredible response to this and so much useful feedback Will continue refining this and release publicly soon! DM me if you want early access, I am doing my best to get to everyone!

Tried to set up the @withoneai integration for GHL and it didn't work (integration id doesn't exist). Trying to get in touch with the team but the verification flow for their Discord server doesn't work (connection reset on redirect page).

@danshipper Found the repo: github.com/EveryInc/proof… Looks like it stores documents in sqlite. Could be okay if the CLI exposes a search function.

@danshipper Can I run it locally so I can own my own data? Don’t want it in the cloud. Want the data to live in my local workspace.

im writing an essay in proof using codex's in-app browser. i type directly in the document, and codex loops in parallel on the left, collaborating with me in realtime. this is obviously the future:

@fagamericano Ha, I babysit my agents in Claude Code so I can watch them thinking and course correct when needed. I’ve been saying for years that we’ll all become machine herders.

I’m at Google Cloud Next and been telling people who ask me what I do that I run a daycare for AI Agents and the reactions have been: - laughter - “I need to send my agents to daycare?!?!” - “who’s your investor?” - “How old?” (filed under creepy) - blank stares. Someone legit stared at me for a good minute.

@fagamericano @oliver_wmr @steipete I tried multiple sandboxed client agents on one box and ran into resource issues (mainly with indexing memories and sessions for semantic search). Are you running on a large instance? Dedicated hardware? How do you deal with CPU contention?

@oliver_wmr @steipete One gateway. 300 workspaces (and their agents).

I’ve been running an “enterprise” openclaw install with 300 users @steipete (gcp, gke, gvisor sandbox, read only workspaces, etc) connected (read only) to new relic, github, google (with its own restricted account) and notion. It has changed how many teams work. Ops has been diagnosing customer issues in record time. It grabs logs from new relic, checks business contracts in notion and checks the repo to find where the transaction failed and why. “This customer support agent made this mistake here”. SecOps built a skill that does a RFC review (change management) and helps engineers address security issues before submitting the RFC. Devops is creating a similar one with infrastructure contracts in mind. We were recently at cloud next as a team checking out things, we added an agent to the channel to help coordinate the offsite. It helped coordinate restaurants, bars, activities, it ran polls, selected places, provided everything we needed in terms of directions and bookings. We just enjoyed the con and would get random Slack pings “choose dinner options close by to Mandalay!”. Some things are a bit… hacky still that I’ve had to put together to reconcile workspaces of users but so far I’ve automated the deployment to GKE and updating the PVC and even rotating secrets with SecretManager (injecting them at pod startup). The keeping it away from PII, read only to prod logs, sandboxing execs and giving it very reduced scope does help some of the anxiety of it doing catastrophic things It certainly is fascinating opening up all the user sessions as an admin and seeing all their use cases in every department. How the engineers use it, HR, Ops, etc.