Eventually, there will be sets of tools for various platforms, e.g., filesystem tools for Node.js and Bun, and a set of tools specifically for building agents that use certain sandbox providers.
Claude Code et al. are amazing, but they are not that great to build on top of them. That's why I created OpenHarness: a code-first, composable, and configurable SDK for building powerful, capable AI agents deeply embedded in existing applications.
Can you tell my why you think I should choose a project started yesterday[1] by one person (driving an ai), versus something like ADK, developed by many people and backed by Google?
Fair question! ADK is a great framework for building agentic systems. OpenHarness is meant to be a set of higher-level primitives that can be used to quickly run more general and capable agents, and can easily work with existing AGENTS.md or agent skills standards. Basically, if you automated a task with Claude Code, and now you want to be able to *easily* run it programmatically. Or if you want to integrate an OpenCode-like agent into your application. OpenHarness is obviously new, but it also doesn't reinvent the wheel - it's built on top of Vercel's AI SDK!
Sutro | Senior Full-Stack Engineer | REMOTE | Full time
At Sutro, we're building the next generation AI-native no-code platform. Our code base is mostly written in TypeScript, with React (Native) for the frontend, with a particular focus on complex agentic workflows and many patterns from the fields of compilers and the theory of software languages.
We are looking for an experienced full-stack engineer with experience in architecting agentic workflows, building tasteful UIs, and knowledge in the field of compilers and software architecture theory.
Our primary location is Oakland, CA, and as such, our core working hours with everyone online are 8 am-11 am California time. Feel free to hit me up with questions!
I find it weird that the only sample application built with it is the old Wunderlist. Wunderlist 2 - which was released a few weeks ago - was developed native for each platform.
Perhaps it may seem so. That said, the original Wunderlist was likely the most infamous use of the original technology that reached millions of users.
It's no slight against its makers to have gone purely native but the fact is that most companies developing a todo app do not have millions in investment or the sort of native programming capacity to throw at this kind of project.
TideKit is a solution that is going to get you on all platforms quickly and its only improving over time. As you might be aware, the new Wunderlist 2 seemed to have lost its linux support as a result of their recent shift.
We're hoping to show something quite soon that shows a speed and quality of development that will have most wundering why they transitioned to native.
I think what the guy you're replying to was saying is that the entire page seems to be about Wunderlist, when they no longer use this SDK, and their use of it isn't really that complex.
Perhaps it would be better to show some actual users with a carousel type thing?
Looks very promising. I think there is still no good tool to manage all your links correctly. However, I have some ideas for you:
- Sharing lists (so that they show up under your list) with teams
- Mobile integration if this is possible
That is exactly what annoys me when looking at the website of a new framework. Often you are completely impressed by how simple it is to set up a blog but while building a larger application you realize that it is not suited for your project.
They have good and well written articles that cover local, national and international subjects and they have a nice website and a very good mobile website. And their office is just next to ours :) Nice to see them on the yc frontpage :)
Eventually, there will be sets of tools for various platforms, e.g., filesystem tools for Node.js and Bun, and a set of tools specifically for building agents that use certain sandbox providers.
reply