I hope you're all doing automated pen-testing of your apps before the bad guys do!

@mojombo thank you for the chat back then. Your entry is published and I hope that we'll meet again. whysfox.es/entries/252-mojom…

I think the next big AI unlock will be powered by agents talking to each other. Your agent talking to mine. My agent talking to a vendor's agent. My agents talking to each other. That's the topic of the next #AgentCribs meetup. Join us! agentcribs.pwv.com/events/oc…

Our hackathon winners got their Mac Minis signed by the creator of OpenClaw and the cofounder of GitHub and investor at PWV. This one went to Michael Starkenberg, whose agent Rundown led real usage out of 48 entries. @sodio @mojombo @steipete @aiworthusing

✅ OpenClaw 2.0 is out. ✅ It's almost the weekend. Time for a hackathon! The @aiworthusing x @openclaw hackathon is happening now. Next week I'll select the winner and fly them to SF for @PWVentures AgentCribs event. Plus hand them a Mac Mini. 🍎 luma.com/zhkhsnpa

> Imagine what happens when millions more people can build robots, medical devices, instruments, vehicles, satellites, toys, tools and things none of us have thought of yet. More builders is more better!

Want to win two Mac Minis and a trip to SF? Our AgentCribs community is hosting the AI Worth Using x @openclaw hackathon: give an agent a real startup job and get real people using it. Winner demos onstage with ME at the event in SF on Oct 6. Sept 22–28. Details @aiworthusing

1.1 Billion tokens in @CommandCodeAI came out to like $15.30 ($10 GOAT plan + $5.30 in API usage). Got tons of work done with it. I don't know how much better of a deal you can get! Like this isn't an ad, I haven't been sponsored, I'm just impressed!

Today we’re releasing our methodology for evaluating model routing with interactive benchmarks, which represent agent cost accumulation better than static benchmarks do. Across leading benchmarks, we achieve Pareto-dominance, exceeding Opus xhigh quality at 20–80% lower cost.

Today we’re announcing Not Diamond Code, the world’s most powerful intelligent model router for long-horizon coding agents. Not Diamond works with any gateway or harness, including Claude Code, to select the best model and reasoning effort for each step, reducing costs by 20-65% without impacting quality.

y’all @CommandCodeAI has massive capacity for Kimi K3. And sub starts at just $1 Go with 10x free credits. New? Command Code is purpose built for open models. Tool repairs, cache repairs, and phenomenal harness engineering. Currently #1 coding agent harness on OpenRouter.

Today is a big day for Routine! 🎉 After 5 years of building, we're finally unveiling Routine AI. Routine AI transforms your workspace — tasks, calendar, notes, projects etc. — into an AI-powered assistant with voice commands (a.k.a super Siri), AI agents, and automations.

Today we're open sourcing Bolt Slides. Now any agent (Claude Code, Codex, Bolt) can make slides you couldn't even imagine before. What does that mean? Take a look 👇

Command Code just crossed 15K paying customers! 🥳 It took us 60 days to get here. 3T tokens. We added yet another $1M a month, and now doing $2M run rate!! July will be crazy, GUI, SDK, ACP, v1 GA, BYOK, major new plan, and two new products launching. Command Code is quickly become the best most used coding agent harness for top open and closed models. We took an early bet on open models when all other agents were only focused on closed frontier models. Our research has led to the discoveries and meaningful inference engine improvements like: 1. tool call repairs (make open models faster and outperform closed by improving their tool calling ability, now doing 1M repairs per 1T tokens on most open models) 2. /design skill (a super well researched bundled skill built by real designers which is an easiest way to deslop AI's design "design repair" and improve just about any design surface) 3. cache thrashing repairs (saving tens of thousands of dollars in costs to our customers with 97-98% median cache across the top 5 models, gateway fallbacks are a bad idea, most gateways are costing you money, strong prefix building and cache management can lead to massive infra/cost improvments) … and a couple of amazing new things we're cooking, about to launch this month! We're also hoping that alongside v1 launch we'll be able to open source Command Code, and make it fully hackable, extendable, and support BYOK, with Codex, Claude, Grok, and more subscriptions working natively. Super easy to get started with $1 Go plan with 10x credits, literally $10, extends to $40 on DeepSeek and other deals as we're not overcharging like all our competitors. Massive roadmap ahead, thank you for all y'all's support. Let's go! 👊

Command Code is the first & atm only coding agent that supports Muse Spark 1.1 by Meta. Shipped on the launch day. Excellent score for the cost on Command Bench. 🔹Context: 1M 🔹Input: $1.25/M 🔹Output: $4.25/M 🔹Cache Read: $0.15/M @finkd @alexandr_wang thanks for sharing.

Ona (formerly Gitpod) is joining @OpenAI, bringing their secure cloud execution and async agent orchestration tech to further enhance Codex! Five years ago, I sent @jolandgraf a DM on Twitter after seeing how rad Gitpod was for spinning up cloud development environments. It seemed clear to me that devs needed better ways to manage multiple environments for different projects, especially in the enterprise. I made my first angel investment in 2021. As I got to know Johannes better, I realized what a special founder he is. Always asking the hard questions of himself and his team. Relentlessly pursuing product/market fit. Extensive communication to the team and investors through Notion docs. In 2022, I led their Series A at @PWVentures, a first for me, and one of the biggest checks I've written to a startup. Over the last four years, Johannes and team have been through several mini-pivots, always adjusting to the market. As AI has taken over mindshare, they made their biggest pivot yet, rebranding and leveraging their core tech to bring robust AI tooling/orchestration via secure cloud development environments to the enterprise. Ona's technology is years ahead of competition, and I'm really glad to see it land at OpenAI, where millions of devs will get to experience it make their lives better. Congratulations to @jolandgraf and the whole Ona team—a well deserved success, and a bright future for you and the product!

I got a new MacBook Pro recently and the trackpad is being weird: When I two-finger scroll, stop, then release my fingers to scroll again, the content I'm scrolling will occasionally jump (like a phantom scroll input). Sometimes a little, sometimes a lot. Never once had this problem on my previous MacBook Air. Anyone seen this? Can't find anyone else on the web with this problem. It's killing me!

Open models are having their moment and Command Code is best harness to use them.