CraftBot with Living UI - Grow your own software that is alive.
by•
Living UI is a brand-new system that lets CraftBot (general AI agent) build, import, or evolve custom apps/dashboards that live inside CraftBot itself. The agent stays context-aware of the Living UI's state and can read, write, and act on its Living UI directly. A Living UI is never "finished". Ask CraftBot to add features or redesign a view as your needs grow.
Living UI turns software from something users buy and adapt to into something CraftBot creates and adapts around them.
Replies
CraftBot
For context, CraftBot is a self-hosted, proactive AI agent that can control a PC and take actions for you.
Today, most software, dashboards, apps, and subscription tools are static. When you need a new feature, you email the developer. If you are lucky, it ships six months later. If not, you may never hear back.
Therefore, we are introducing a new concept: 🌱Living UI🌱.
Living UI lets CraftBot build, import, and evolve custom software, apps, and dashboards directly inside the agent. The best part is that CraftBot stays context-aware of the Living UI’s state. It can read, write, update, and act on the interface directly.
Need a Kanban board with a general AI agent built in?
A CRM tailored exactly to your workflow?
A company dashboard your agent can actually operate?
You can spin all of these up as a Living UI. There are three ways to create one.
1. Build from scratch. Just describe what you want, and CraftBot generates the backend, API, and UI, then iterates with you.
2. Install from the marketplace. Use ready-to-use apps built by the community.
3. Import your existing project or GitHub repo. CraftBot converts it into a Living UI and integrates itself into it.
A Living UI is never finished. You can always modify it by telling CraftBot as your needs evolve. So instead of static tools, you get software that grows with you.
Try CraftBot and Living UI today. Build and customize your own Living UI, and stop relying on subscription tools that were never built to fit your needs perfectly. 🟠🌱
PicWish
@tham_yikfoong what happens if local machine runs out of storage for logs/memory? does it have a way to auto-prune old events?
CraftBot
@mohsinproduct There is a logic that prunes logs and MEMORY.md file with respect of their file size, but it does not consider the storage of the local machine! Generally they wouldn't exceed 50MB. Great question!
Looks amazing. How good is it with integrations? I am thinking about connecting it to some existing Notion databases to make use of it as internal dashboards.
CraftBot
@nelson_milla Notion is a part of our existing external integrations. You can easily set it up using OAuth or creating your own notion app.
You can ask the agent to create a living UI connected to your Notion account and fetch data in real-time in order to keep the living UI and notion continuously in sync.
quick question on the "Build from scratch" flow how does it handle complex API integrations? if i want my Living UI to pull data from multiple third-party tools, can CraftBot set up those connections too? checking the demo now.. @tham_yikfoong
CraftBot
@tham_yikfoong @vikramp7470 Craftbot has a list of external tools and we're continuously adding more. These external tools are provided to the agent with a simple interface. They allow proactive use and can be easily integrated into the Living UIs without any extra work. We are continuously adding more external tools but other than these, yes the agent can integrate any API needed (you might need to set up API keys though).
The 3AM memory consolidation is the most interesting architectural decision here. Love it. Most local agents treat memory as a simple append log (which means the context gets bloated or stale fast). This one is different.
CraftBot
@artstavenka1Â Glad you like it! Yes, we also prune it periodically, removing and merging memory depending on the context.
How CraftBot connects different Living UI seems really cool use of agentic workflows. So, I am curious if it is mainly one agent doing everything, or do you use multiple agents with different roles that work together in some kind of orchestration?
CraftBot
@rupeshpandith It is actually something we have been considering to improve. Currently, the main agent create Living UI with a creator skill. However, having multiple sub-agents can save more tokens and stuff. Please look forward to our future release! This is the direction we are heading.
Flowtica Scribe
The agent era is reshaping the interface.
TUI is one direction, and it makes more sense for devs. But I don’t think agents remove the need for UI. Visual interfaces are still one of the best ways to compress state, context, and actions into something humans can understand quickly.
@CraftBot’s Living UI pushes this in a useful direction. The interface is no longer a fixed app you buy and adapt to. It becomes something the agent can build, read, modify, and operate based on the current task.
This points to a different software shape: not one-size-fits-all dashboards, but interfaces generated around the immediate job, the user, and even the current moment.
For agents, UI may become less of a destination and more of a live surface for action.
minimalist phone: reduce your screentime
Wishing good luck with today's launch :)
CraftBot
@busmark_w_nika Thanks Nika!
CraftBot
@lakshminath_dondeti Let me answer both of your questions here. The UI change when you ask CraftBot to modify them, and they are device responsive so it adapt well on the CraftBot browser interface whether it is opened in web or mobile. It can be both dashboards/apps, since dashboard is kinda the subset of apps.
The shift from reactive to proactive is a much harder design problem than most agent products acknowledge. Reactive agents fail silently, you just don't get a good answer. Proactive agents fail loudly, wrong actions on your behalf, noise surfaced as priority, the things that actually mattered missed while chasing the things that didn't. Curious how CraftBot calibrates when to act vs when to ask, and whether that threshold adjusts based on how often the user overrides it.
CraftBot
Fair push. Being honest: the roadmap still lists Proactive Behaviour as Pending, so today the answer is mostly the "with approval" step in the agent loop; closer to reactive-with-a-confirm than real calibration.
Where I think it lands, given the current architecture: the Skill Manager and Action Router are the right place for this, not a global confidence knob. The axis that matters is reversibility, not certainty; a Skill that drafts into Gmail and one that sends are different decisions regardless of how sure the model is, so the threshold should live in the skill manifest. Override-as-signal fits naturally on top of the Memory Manager (it already consolidates events at midnight), but probably as a gradient edit-before-send is soft, delete-and-rewrite is hard, disabling the skill is structural rather than one counter feeding back into a single threshold.
Still early, and open to where you've seen this work.
mailX by mailwarm
Most tools still force users to adapt to fixed interfaces and systems, but the idea of an AI agent staying true to context and evolving the UI as needs change feels much closer to how people naturally work. Really interesting direction.