For you, your agent, your coworker and their agent. It holds the team's critical know-how, research, decisions and data. But it's not a dead storage. It's a workspace that makes the context workable for humans as well as agents.
@ben_stephenson2 this is always a challenge. The brain is made to self-evolve and clean up. Project are made to be messy. We plan to have more agents soon so you can setup linting agent that goes over workspace and find things that are duplicate, old, and maybe not relevant!
@ben_stephenson2 yep we are working on a system that is going to help with this. We call it "gardener". It will run every morning and tell you what can should be merged / deleted / is out of date.
@ben_stephenson2 we are trying to find balance as AI is still not in the state of handling the context on its own, and we believe it is very important to keep human in the loop. we are working on "gardener" workflows which should make agent resurface relevant stuff and support this flow in more productive way
@ben_stephenson2 We are thinking about it this problem a lot and we are currently working on something we call a "gardener agent", that automates and simplifies this process.
Report
I built something like this internally for our team - literally called it our Brain. The idea was they can use that context in Claude though and build skills against it, etc.
Is that possible, or can you only execute tasks within Kanwas? I love the idea of a Brain we can all visualise & collaboratively maintain/iterate in here but then use in claude.
@tom_rudnai1 Yes definitely. You can use Kanwas as the single source of truth for context that the whole team expands and edits and use it with other AI tools.
@tom_rudnai1 exactly! updates work in both directions. you can bring anything in kanwas and iterate on all your signals, and also local/coding agent can pull the latest context and work with it
@tom_rudnai1 Yep, that’s the beauty of Kanwas being backed by simple file formats like Markdown. You can easily import and export everything.
Report
love that it’s fully editable. my biggest fear with "ai brains" is when they hallucinate a decision we never actually made. being able to step in and refine the context keeps the agents from going off the rails. @johancutych
@priya_kushwaha1 exactly, we made it fully transparent and actually pleasurable to write and edit files. Cause that's how you keep the context up to date and tight!
@johancutych@priya_kushwaha1 Yes, having full control over the context is one of the most important features of Kanwas. In the future we would like to explore making even the thinking process and the context references in one conversaion visiable.
@priya_kushwaha1 yep one of the biggest problems with current tools. And why we are making kanwas. Its made for introspecting your agents brain. This is how you get 10x output.
@priya_kushwaha1 Thank you Priya, exactly on the point! The core we are focused on is keeping the human in the loop, and with interface supporting you trust your output, by knowing and controlling how you did it.
I am glad this is not another closed workspace where everything disappears into a proprietary database. Markdown/YAML plus Git history makes the product much easier to trust.
@mahin_makkhy This kind of interoperability was very important to us. We didn't want to vendor lock users to some proprietary format but rather let them work on markdown files.
Nice side-benefit of this is that you can use our CLI tool to export and import files easily
@mahin_makkhy I'm glad it resonates with you. We belive the era of heavy close, proprietary tools is comming to an end. Open source and open standards / formats are now important more than ever and we've been building Kanwas with this in mind since the first day.
Great idea! I have been struggling with all the rigid chat UIs out there that try to fit your worflow into a linear conversation. But that is not how brainstorming usually works. It is messy and canvas is a great solution to that!
@zuse yep collapsed outputs are hell, I've always liked canvas interface for any creative and iterative work. messy middle is the new moat!! thanks for shouts Tom!
@zuse Thank you! Team brainstorming session with Kanwas over Google Meet is one of our most favourite use cases. We all have the shared context, there is a visual reference to reason about, and we can ask the agent to check any facts or do some additional research as we go.
@zuse Thank you, Tomas! Canvas is a big part of the product, and our approach is to provide a more neutral space for brainstorming and creative thinking.
@dan_tatar There are over 1000 native connections, and a native CLI tool that exposes the raw file system. I would say that the connection possibilities are only limited by imagination.
@igor_kuznietsov I would call it well oiled machine :) Folders that have nice compact human readable context that AI can use to make great decisions in. You will feel significantly more productive here then you would be in Claude chats and other linear tools.
@igor_kuznietsov That's a good question. In the ideas scenario, where you invest a bit of time into actively managing the context, it grows and compounds. You will have a lot of general context about the company, the market, etc. and also you will capture the specific work on specific problems like crafting marketing copy, PRDs or strategic documents, which both helps you get better reasoning and outputs from your AI agents.
@igor_kuznietsov Hey Igor, Kanwas is updating its memory accordingly and keeping the log decision. Also on the side we are preparing the gardener flow which will support you with old, left artifacts, by resurfacing them and cleaning by date and relevance.
How do you stop the canvas from becoming another messy place over time? Is there an agent workflow for cleaning up old context and decisions?
Kanwas
@ben_stephenson2 this is always a challenge. The brain is made to self-evolve and clean up. Project are made to be messy. We plan to have more agents soon so you can setup linting agent that goes over workspace and find things that are duplicate, old, and maybe not relevant!
Kanwas
@ben_stephenson2 yep we are working on a system that is going to help with this. We call it "gardener". It will run every morning and tell you what can should be merged / deleted / is out of date.
Kanwas
@ben_stephenson2 we are trying to find balance as AI is still not in the state of handling the context on its own, and we believe it is very important to keep human in the loop. we are working on "gardener" workflows which should make agent resurface relevant stuff and support this flow in more productive way
Kanwas
@ben_stephenson2 We are thinking about it this problem a lot and we are currently working on something we call a "gardener agent", that automates and simplifies this process.
I built something like this internally for our team - literally called it our Brain. The idea was they can use that context in Claude though and build skills against it, etc.
Is that possible, or can you only execute tasks within Kanwas? I love the idea of a Brain we can all visualise & collaboratively maintain/iterate in here but then use in claude.
Kanwas
@tom_rudnai1 Yes definitely. You can use Kanwas as the single source of truth for context that the whole team expands and edits and use it with other AI tools.
Kanwas
@tom_rudnai1 good to see you here Tom!!! Actually we built it so you can run any agent through CLI :)
Kanwas
@tom_rudnai1 exactly! updates work in both directions. you can bring anything in kanwas and iterate on all your signals, and also local/coding agent can pull the latest context and work with it
Kanwas
@tom_rudnai1 Yep, that’s the beauty of Kanwas being backed by simple file formats like Markdown. You can easily import and export everything.
love that it’s fully editable. my biggest fear with "ai brains" is when they hallucinate a decision we never actually made. being able to step in and refine the context keeps the agents from going off the rails. @johancutych
Kanwas
@priya_kushwaha1 exactly, we made it fully transparent and actually pleasurable to write and edit files. Cause that's how you keep the context up to date and tight!
Kanwas
@johancutych @priya_kushwaha1 Yes, having full control over the context is one of the most important features of Kanwas. In the future we would like to explore making even the thinking process and the context references in one conversaion visiable.
Kanwas
@priya_kushwaha1 yep one of the biggest problems with current tools. And why we are making kanwas. Its made for introspecting your agents brain. This is how you get 10x output.
Kanwas
@priya_kushwaha1 Thank you Priya, exactly on the point! The core we are focused on is keeping the human in the loop, and with interface supporting you trust your output, by knowing and controlling how you did it.
Slazzer
I am glad this is not another closed workspace where everything disappears into a proprietary database. Markdown/YAML plus Git history makes the product much easier to trust.
Kanwas
@mahin_makkhy This kind of interoperability was very important to us. We didn't want to vendor lock users to some proprietary format but rather let them work on markdown files.
Nice side-benefit of this is that you can use our CLI tool to export and import files easily
Kanwas
@mahin_makkhy .md files are way to go!!
Kanwas
@mahin_makkhy True! Markdowns are the way to go, and from our learnings agent works the best when it lives on top of it.
Kanwas
@mahin_makkhy I'm glad it resonates with you. We belive the era of heavy close, proprietary tools is comming to an end. Open source and open standards / formats are now important more than ever and we've been building Kanwas with this in mind since the first day.
CalendarPipe
Kanwas
@zuse yep collapsed outputs are hell, I've always liked canvas interface for any creative and iterative work. messy middle is the new moat!! thanks for shouts Tom!
Kanwas
@zuse Thank you! Team brainstorming session with Kanwas over Google Meet is one of our most favourite use cases. We all have the shared context, there is a visual reference to reason about, and we can ask the agent to check any facts or do some additional research as we go.
Kanwas
@zuse Yep we love to use for these creative sessions. There really is no good tool to do this in.
Kanwas
@zuse Thank you, Tomas! Canvas is a big part of the product, and our approach is to provide a more neutral space for brainstorming and creative thinking.
Nice, is it synced with any tools in particular?
Kanwas
@dan_tatar There are over 1000 native connections, and a native CLI tool that exposes the raw file system. I would say that the connection possibilities are only limited by imagination.
Kanwas
@dan_tatar we have all connections out there, plus cli as the input!
Kanwas
@dan_tatar Yep we are using composio.dev for all connections. There are more ten 1k+ and they work much better then MCPs
Kanwas
@dan_tatar we have 1k+ connections and it actually work wonders!
What does a Kanwas workspace look like after a project has been running for a month?
Kanwas
@igor_kuznietsov I would call it well oiled machine :) Folders that have nice compact human readable context that AI can use to make great decisions in. You will feel significantly more productive here then you would be in Claude chats and other linear tools.
Kanwas
@igor_kuznietsov That's a good question. In the ideas scenario, where you invest a bit of time into actively managing the context, it grows and compounds. You will have a lot of general context about the company, the market, etc. and also you will capture the specific work on specific problems like crafting marketing copy, PRDs or strategic documents, which both helps you get better reasoning and outputs from your AI agents.
Kanwas
@igor_kuznietsov Hey Igor, Kanwas is updating its memory accordingly and keeping the log decision. Also on the side we are preparing the gardener flow which will support you with old, left artifacts, by resurfacing them and cleaning by date and relevance.
Kanwas
@igor_kuznietsov Long time no see man! Actually it looks tight and clean cause the brain self evolves and project are easily cleaned up by agents!