it worked, but was hard to tell what agents were doing, why something failed, or whether a workflow was actually progressing.
We thought it would be more interesting to treat agents as long-lived workers with state and responsibilities and explicit handoffs. Something you can actually see and reason about, instead of just tailing logs.
So we built Clawe, a small coordination layer on top of OpenClaw that lets agent workflows run, pause, retry, and hand control back to a human at specific points.
This started as an experiment in how agent systems might feel to operate, but we're starting to see real potential for it, especially for content review and maintenance workflows in marketing. Curious what abstractions make sense, what feels unnecessary, and what breaks first.
These are all proto orchestrators. No one has discovered/converged on what agent orchestration actually looks like.
Other projects include conductor.build, gas town (infamously), and others.
Another layer of abstraction in the infinite castle of computer science.
please hold for the non-existent cdn while we download 372 fonts
Whereas the former see things talking to each other across boundaries with a minimal set up, the latter might look at it and see/say: "hard leaky abstractions"/"Poor security model"/"High $/task", you can't you build/sell it.
Where the latter crowd might exclaim: "Look my slack bot can confirm with my doctor when calendar appointment is near. SaaS tools are dead!". The engineering/dev/product are more sceptical and they are right to be so.
It's also an atomic bomb of a security hole waiting to explode.
Really makes me want to be here.
I polish stuff before promoting because I'm averse to reputation damage, but every day that I see people not doing that and getting upvotes anyhow makes the practice harder to justify.
Kanclaw
Clawban
Clawboard
Clawlo
Clask (Claw-task)