gm world changers! It's time to settle in for another issue of the Leaderboard. In today's digest, we're diving into a new deployment framework from Meta, a unique method of avoiding distracting sites, and a Dreamweaver for the AI era. Let's dive in.
Commit to the stack
Llama Stack: An open-source framework for simplyifing AI deployment.
Llama Stack is interesting as Meta’s first-party push into open AI ecosystems. It’s a solid choice if you’re already deep into open models, offering a comprehensive setup for local and cloud deployments. However, it lacks flexibility for integrating or migrating from proprietary services, which could make adoption trickier in the short term. Still, for those fully committed to the open model stack, it’s a promising option.
Last resort productivity
TabBoo: Add random jumpscares to sites you're trying to avoid
BOOO!!! That's kinda how it feels like with Tabboo when you're wasting time on the internet. It's honestly such a smart idea, to have something jump scare you into going back into work. Kinda feels like when you know your parents or boss are going to walk in the room so you should probably switch tabs...but don't... only this helps you avoid the consequences of being distracted. Honestly, awesome job makers. Now brb while I install this on my wife's computer for amazon and rhode.
Building has never been easier
co.dev: Turn ideas into full-stack appsÂ
Trying co.dev brought me back to building my first website in Dreamweaver, watching code update in real time. Fast forward 20+ years—I built a Chrome extension with ChatGPT, but troubleshooting and deploying changes was a headache. Co.dev feels like a magic portal for non-technical users, letting me build web apps with natural language and see updates instantly. If only it also handled the admin side of launching to the web!
Your agents stopped answering and started doing
Your agents call tools, write to databases and move money, so a wrong answer is now a wrong action, which is a considerably worse meeting. Arcjet runs policy before every LLM call, tool call, query and API request: prompt injection, sensitive data, bots, rate limits, spend quotas, and your own rules in Rego on top. It reads the OpenTelemetry you already emit, so switching it on is one environment variable, no code changes and nothing to deploy. Each check also reads the earlier steps in the same workflow, so an agent's tenth action gets judged against the first nine and not on its own. JavaScript, Python and Go SDKs all at 1.0, agent framework guards included, free tier to start on.
Daily Top Products
