Launched this week
Viberia
Command AI agents like you're playing Civilization
315 followers
Command AI agents like you're playing Civilization
315 followers
Do you like your Claude/Codex pet but wish you had a zoo? Viberia is a spatial command center for your AI agents. Your whole AI org lives on an isometric map, status icons show who’s blocked, who’s asking, who’s done. Zoom in to chat with anyone, pick any model, bring your own keys. Agents collaborate, build their own little teams, and pick up new skills as they go. Docs, terminals, browsers built in.
Viberia
Hey Product Hunt!
I’m Emre, Viberia’s solo builder.
I spent the last year drowning in terminal and IDE windows trying to keep up with my own agents. MCPs, subagents, skills, hooks, modes and more.
I tried Conductor, Claude Squad, Gas Town and others. They all felt the same after a while, just a long list of things waiting on my attention.
So, I went the other way.
What if managing your agents felt like playing a game of Civilization or SimCity? You zoom out and see your whole AI org on an isometric map, with status icons telling you who’s blocked, who’s asking a question, who’s done. You zoom in and chat with anyone. The agents collaborate, hand work off to each other, and over time they build their own little teams and pick up new skills.
A few things worth knowing:
- The harness is built around teams/buildings, these are agents that work together (and/or in a sequence) to deliver what you need (e.g., write PRD first, write code later, and then review). I saw my own productivity and output increase significantly when I started using agents in a sequence, and this is one of the core differentiators of Viberia.
- It’s free and will stay free; you just need to bring your own subscription and/or API keys. It works with Claude, GPT and Gemini.
- Docs, terminals, browsers are all built in. No app-switching.
- It’s a Tauri app. Roughly 8x less energy than the closest competitor I benchmarked, so you can actually use it at a coffee shop on battery.
The product is still in early alpha. There will be bugs and missing features. Please tell me about them. I read every comment, and I’ll personally chase every bug and feature request. My goal is to make Viberia the most efficient, capable, and fun agent harness out there.
Cheers,
Emre
Congrats on the launch, Emre! I am really curious about the architecture under the hood. As these agents collaborate, hand off work, and pick up new skills, is there a knowledge graph getting formed on the backend from each LLM to manage their shared context and relationships?
Viberia
@sammy_rawat Thanks Sameer!
There is no default knowledge graph that tracks their collaboration/relationships, but you can add yours in Viberia! There is an internal text-based memory system, and the agents decide when to use it. In my experience they keep track of their relationships, but you could give them more specific guidance on this through the system prompt.
If you want to plug in your own knowledge graph, drop any memory/KG MCP server into the agents and tell them to use that whenever they collaborate with each other. You can set this per-agent, per-team, or per-project. I didn't want to bet the architecture on one memory paradigm, so I left the choice to you.
Thanks for the question! I honestly hadn't seen this as a gap, but you're making me reconsider. Maybe I should build collaboration tracking deeper into Viberia.
@emre_barut great, you have really thought deep on the architecture and kept it modular, nice to see that.
Viberia
@sammy_rawat Thanks! @Obsidian was a huge inspiration here.
I'm going for "you can bring anything in, we are not locking you in or out, you can export everything if you really want to, but you won't because this will be the best way to do things".
This is a great initiative, just wanted to know if you thought of adding an additional feature where different agents remember your codebase and you don’t have to give the context again and again. So my main thought is if we are working on a project so you can divide these agents into teams say frontend, backend team and all the agents works in accordance to each other. If changes are made then all the teams are informed so that every agents works accordingly and there are less coding conflicts. This feature can also save you credits usage.
Viberia
@partha_gautam Thanks for the great comments and the feedback!
That's exactly how I've been using Viberia to build Viberia. I keep docs, CLAUDE.md (+ AGENTS.md) and README.md files for different parts of the codebase, and when agents start working on a section they read these to get the full context. Saves a TON on tokens, as you said.
I haven't added this as an explicit feature yet (and I probably should), but the best way to set it up today is to spin up a documentation team and literally give them this task. Then, agents/teams can use that info properly.
For parallel work I use git worktrees. There's a building in Viberia called "Worktree Forge" that sets these up for you. It takes two clicks to spin one up, or you can ask an agent to do it (which is how I usually do it). Each team works in a different branch, and the Merger agent in the Forge handles the merging while staying aware of the docs and potential conflicts.
If you want to take this one step further, use the Linear building (assuming you track issues in Linear; you can use a different provider, just set up your own custom building with the right MCPs). Ask the agents there to come up with 6-7 tasks you can tackle in parallel. The Issue Resolver agent will set up the worktrees, brief each team on what they need to do, and tell them what conflicts they might hit with other ongoing tracks. The worktree agents get back to you when they have a plan or an implementation, and then you OK it or ask for improvements. Makes for a really fun iterative development loop. I work with 6-10 branches in parallel using this setup (see the screenshot below of a section of my git commit graph).
pretty cool concept@emre_barut!
Curious what are your thoughts regarding visualization/ui/ux for the management of tasks and projects rather than teams and agents? I feel like grooming tasks of different stages (e.g. developing prototypes to get more clarity, or, when enough clarity has been reached, making sure the architecture/plan is good) is a place where bottlenecks are moving into (besides the actual pipeline of how a team of agents handles work)
also...pretty fun to see the Lion as the chief/mayor in the pictures, got a bit of zootopia vibes which I thought was lovely
Viberia
@luis_hernandez23 Thanks Luis!
The way I've been using Viberia, the building is the task. Buildings don't have to be permanent teams. Half the time, mine are task-specific: I spin up a building for a feature or component, fill it with the agent roles that make sense for the flow (planner, developer, reviewer, etc.), and tear it down when the work is done. Each agent inside ends up handling a specific section of that flow, so an agent's identity is really "doing X (the agent role) for component Y (the building name)".
So if you want lifecycle stages, you can literally model them as buildings: a "Prototyping Lab" for exploration, an "Architecture Council" once you have enough clarity, a "Build Team" for execution. You can wire them up so work flows between them, with the Linear building plugged in for the formal ticket layer if that's how you track things.
That being said, you're right that there's no first-class "task overlay" on the map today. The task identity is implicit in the building, not something you can see and move around independently. There is probably a way to use the spatial layout to surface that better, definitely something I should think more about.
On the Zootopia vibes, that's probably been subconsciously drilled into my brain by someone at home who won't stop talking about it, especially since the second one came out..
Viberia
@luis_hernandez23 Thanks! This is very similar to how I was approaching it, but I didn't do a great job explaining this in the presentation. Viberia comes with 8 built-in buildings. They are either (i) flow or (ii) subject driven. For instance, the Council building does Karpathy's LLM council, where multiple models answer a question and then they rate each other (that's flow-driven). The Knowledge Base building/team is built for going over your Obsidian vault (or other knowledge base); agents focus on tasks like adding new content, organizing current content, or helping you fetch content from your library (that's subject-driven). You can also set up your own custom buildings, and the Chief of Staff agent has multiple skills that can directly do this, by going over your needs and the current structure of your codebase/project.
I don't have an inspector agent (that's a great idea!), but that would be a very easy addition. I also wanted to have a security building that can check for funny stuff (e.g., prompt injections or models going haywire). All of this is buildable within the current interface.
> And congrats on the fam growing bigger! gotta enjoy them as much as possible (and have the viberian agents working in the background 🏗️)
Thank you! I can't overstate the usefulness of agentic development when you have kids. You set up tasks, then you monitor from a playground bench while they're on the swings.
@emre_barut that makes sense!
Also this approach made me think there's an opportunity to use Viberia to teach and learn about multi agent systems. E.g. in 4x games like civilization, after playing for a while, one gets to develop a sense of what's a good strategy / use of resources vs a not-so-good one, becoming more efficient and effective over time.
I imagine there could be some sort of competition (even against the AI initially) to build projects / create products, etc... [using Viberia] that meet a set of rubric criteria (in real life projects this is very open, but for training and simulation purposes one can constrain it).
That's prob a different product geared towards playing and learning (although it leads to building real skill and real things as well)... the foundation of what you have here and the 4X elements made me think about that potential [but maybe I'm stretching the game-like analogy too far...in any event thought I'd share those 3 am thoughts 🌝💭]
Viberia
@luis_hernandez23 These points are way too good, to the point that this is getting a bit uncanny... You are laying out ideas I've been thinking and talking about for a while. Have we met in real life?
I was thinking about using Viberia for teaching/learning about multi-agent systems too. It could be "Scratch for AI agents".
Do you mean this in the sense of setting up a specific rubric, launching multiple agents for the same task, and then maybe doing evolutionary learning (worst ones get discarded, best ones get merged, you repeat)? I've only done this with 2 to 3 teams for handling different issues, but the interface makes it easy to scale up something like this. You can also monitor different teams' token costs, and see what approaches give you a better output/cost ratio.
I think that could and should be merged into Viberia. My initial reason for going for the 4X element was to make sure this would scale up to something that could manage 100-200 agents. In Civ (and others), you run empires on a single screen; if you want to run a digital AI empire, I feel like this is the only interface that's been proven.
Please ping me on Discord, LinkedIn or wherever is best for you. Would love to continue this discussion.
really cool idea-feels very different from usual list based tools? Also I am excited to know how you manage it when there are a lot of agents, doesn't it get messy?
Viberia
@yati_kumawat Thank you!
It really doesn't get messy. It just gets fun.
Spatial placement + status icons helps you contextualize what's going on in different parts of your project. You're not reading 20 chat threads, you're scanning a map (e.g., "ok, 5 agents blocked in the frontend, let me focus there first").
Sometimes I kick off 8-10 agents at once. Instead of feeling overwhelmed I'm just smashing buttons to feed them their next orders. Because it feels like a game, it doesn't drain you the way prompting usually does.
And if you really do get lost, just ask your Chief of Staff ("I'm stuck, I have 10 agents who need responses from me, help" is something I might have said before). With that agent, you don't have to type, you just pick among the options it presents. You become the decision maker, not the prompter.
I've run close to 20 agents concurrently and never hit vibecoding burnout. Just looking around to see who needs help.
Give it a try. If something feels off, please tell me (or your Chief of Staff) so we can figure out how you can happily rule your army of AI agents.
Ahah seems so fun, but how do you connect them ? All by one on Claude ? Or another one ?
Viberia
@thithilacastagne 😀 it is fun!
You can mix and match, you pick the model for each agent. There are also default settings for the type of an agent, so you can go "I want gpt-5.5 for developers and gemini for designers". Under the hood each agent gets MCP tools that allow them to interact with the interface ("let me open a md file and show it to the user") and with each other ("let me message folks in my team"). In that sense Viberia's backend connects whatever model you are running for each agent.
Agents are model agnostic too. They retain their settings (memories, skills etc.) when you change models (or providers) so you don't have to stick with a single model for the lifetime of an agent.
@emre_barut great, really !
Thx for your answer :)
agents pick up new skills as they go' line is the one i'd want unpacked. skills in what sense exactly, fine tuning on local context, tool access expanding over time, prompt templates that evolve, something else? that framing could mean something architecturally interesting or it could be a metaphor for agents just getting better context over a session. which one is it and does it persist across sessions or reset when you close the app