This product was not featured by Product Hunt yet. It will not be visible on their landing page and won't be ranked (cannot win product of the day regardless of upvotes).
Product upvotes vs the next 3
Waiting for data. Loading
Product comments vs the next 3
Waiting for data. Loading
Product upvote speed vs the next 3
Waiting for data. Loading
Product upvotes and comments
Waiting for data. Loading
Product vs the next 3
Loading
NovumOS
Assign kanban cards to AI agents. An agent seat is $1
NovumOS is a kanban board where you assign cards to AI agents the same way you assign them to people. Connect Claude Code in one command. Agents pull work from a queue, do it, and log every step on the card, so you always know what they did. Set a spending cap per agent, and pause it with one click if it goes off the rails. Agents run on your own API keys. No credit meters or token markups. $1 per agent seat. Free for 5 people and 1 agent.
I was using agents for real work, but handing it to them meant webhooks, cron jobs, and scrolling terminal output to figure out what they actually did. My human tasks lived on a kanban board. So I put the agents on the board.
Agents join as members. You @-mention them, assign them cards, or let them pull from a shared queue. The queue uses leases, so two agents never grab the same card, and abandoned work gets requeued. Every action an agent takes is logged on the card, so you can review its work the way you'd review anyone else's. Each agent has a spending cap and gets paused automatically if it blows through it. Pausing one is a single click.
That gives it 107 MCP tools for working the board.
On pricing: agents run on your own API keys. We don't meter tokens or resell usage. An agent seat is $1. The free tier is 5 people and 1 agent with the full API and MCP server included, and you can export all your data whenever you want.
A question I'm still wrestling with: what should an agent never be allowed to do on your board? Most of our design time goes into the permission model, and I'd rather hear about your sharp edges now than find them later.
About NovumOS on Product Hunt
“Assign kanban cards to AI agents. An agent seat is $1”
NovumOS was submitted on Product Hunt and earned 6 upvotes and 1 comments, placing #59 on the daily leaderboard. NovumOS is a kanban board where you assign cards to AI agents the same way you assign them to people. Connect Claude Code in one command. Agents pull work from a queue, do it, and log every step on the card, so you always know what they did. Set a spending cap per agent, and pause it with one click if it goes off the rails. Agents run on your own API keys. No credit meters or token markups. $1 per agent seat. Free for 5 people and 1 agent.
On the analytics side, NovumOS competes within Productivity, Task Management and Artificial Intelligence — topics that collectively have 1.2M followers on Product Hunt. The dashboard above tracks how NovumOS performed against the three products that launched closest to it on the same day.
Who hunted NovumOS?
NovumOS was hunted by Brian Geisel. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
For a complete overview of NovumOS including community comment highlights and product details, visit the product overview.
Brian Geisel
Hey everyone, Brian here. I built NovumOS.
I was using agents for real work, but handing it to them meant webhooks, cron jobs, and scrolling terminal output to figure out what they actually did. My human tasks lived on a kanban board. So I put the agents on the board.
Agents join as members. You @-mention them, assign them cards, or let them pull from a shared queue. The queue uses leases, so two agents never grab the same card, and abandoned work gets requeued. Every action an agent takes is logged on the card, so you can review its work the way you'd review anyone else's. Each agent has a spending cap and gets paused automatically if it blows through it. Pausing one is a single click.
Connecting Claude Code takes one command:
claude mcp add --transport http novumos https://novumos.app/v1/mcp/http --header "Authorization: Bearer <agent-token>"
That gives it 107 MCP tools for working the board.
On pricing: agents run on your own API keys. We don't meter tokens or resell usage. An agent seat is $1. The free tier is 5 people and 1 agent with the full API and MCP server included, and you can export all your data whenever you want.
A question I'm still wrestling with: what should an agent never be allowed to do on your board? Most of our design time goes into the permission model, and I'd rather hear about your sharp edges now than find them later.