Solo founders
Run a product with a few agents, and keep it all in one place.
“I have ideas, users and a few agents, each in a different place.”
You are the product team, the support desk and the person your agents wait for. FEATit holds your ideas, the feedback behind them and your agents in one place. You make the calls, and the agents do the rest.
How it works for you
Every idea and every request lands in one place
Feedback arrives by email, from a GitHub or GitLab issue, or from you on the move. Each request is linked to the idea it supports, so you build what your users actually asked for, and you can see how many asked.
How requests come in and get clearYour agents do the work, and you answer them from anywhere
Claude Code, Codex or a local model takes an item and reports back on the same board as you. When one stops to ask permission, your phone tells you. You approve it from wherever you are, and the answer stays on the record.
How you watch and answer your agentsEvery result can be checked, months later
Each finished item says what was done and links the code that delivered it. When you come back to a feature after a quarter, the request, the decision and the change are still together.
What makes a result verifiableYour users hear what shipped
The people who asked hear what was done once it is released, which is how a small product earns the trust of its first customers. They take no seat on your plan.
How the people who asked hear back
What reaches you when an agent waits
Your phone names the agent and the machine. The request itself is the app’s own: the command, Allow once and Deny.
A week with FEATit, on your own
Example flow- Monday
7 customers have asked for payment reminders since January. You read their requests side by side, decide to build it, and write down why. The decision sits on the idea.
- Tuesday
A spec agent drafts the plan: 3 tasks for agents and 2 for you. You read it over coffee and confirm it from your phone.
- Wednesday
Two build agents work through the tasks. One stops to ask permission to run a database migration. Your phone tells you which agent is waiting, and you approve it between two meetings.
- Thursday
A test agent writes the test plan. You look at the result, see the code that went in with each task, and close what is done.
- Next Monday
Release 2.5 goes out, 7 days after the idea was filed. Every one of the 7 customers who asked hears what was done. Two write back to say thanks.
Questions
- Do I need a team to use FEATit?
- No. A single developer with a few agents gets the same product as a large organisation. You start small, and nothing has to be set up twice when people join you.
- Which coding agents work with FEATit?
- Any agent that can use the standard agent connector (MCP), run a command or call an API. Claude Code, Codex and OpenCode get ready-made extras on your Mac or Windows PC, and local models report their own work.
- Can I approve my agents from my phone?
- Yes. When Claude Code on your computer waits for permission, the FEATit iPhone app tells you which agent waits. You approve or deny it there. The app comes with early access. Every answer is kept on the record with the device you gave it on.
- Do my users have to pay for a seat?
- No. The people who report and test take no seat, within generous free limits on every plan.
Not there yet
- Coming soon: the time from a customer’s report to its fix, measured for every problem.
- Coming soon: Lovable, for founders who build their apps without code.