Comparison
Hosted AI coding agent or self-hosted agent: what changes
Short answer
A hosted agent runs on its provider’s machines. Reqursor Development runs on a Linux server you operate, with its secrets on your data volume and spending caps that you set per project.
| Aspect | A hosted AI coding agent | Reqursor Development |
|---|---|---|
| Where the agent runs | On the provider’s machines | On a Linux server you operate |
| Who runs the server | The provider that hosts it | You, with Docker or a one line installer |
| How you get it | As a hosted service | You install it; there is no hosted version of the product itself |
Where the code and secrets live
A hosted agent runs on the machines of the provider that hosts it. Reqursor Development is installed on your own server, and there is no hosted version of the product itself, only the license server that activates it.
State, encrypted secrets, your license, project workspaces and run artifacts live on a single data volume on that server. That volume is the only thing you need to back up.
Who sets the limits
You set the budgets: caps per ticket, per day and per calendar month on each project, and Autopilot starts nothing new when a cap is hit.
What each cap does when it is hit is covered in Control AI spend.
What you operate yourself
Self-hosting means you run the server. You choose Docker or a one line installer, keep the data volume on real storage, and make sure the tools your checks need are installed.
Settings shows what is new in an update. Confirming one takes a backup first, verifies the signature, lets any running ticket finish its current stage, and restores the backup automatically if a migration fails. On Docker the update itself is a pull and a restart.
When the license expires
An expired license stops new work at the next ticket boundary. It never deletes data or touches your repositories.
The license is checked once a day.
What the daily check carries, and what happens when it cannot be made, is covered in Security sensitive teams.
Good fit
- Teams that want the agent, its secrets and its run history on a server they operate
- Teams that want to set the spending caps themselves on each project
- Teams that can run a Linux server and keep a backup of its data volume
Not the right fit
- You want the agent hosted for you: there is no hosted version of the product itself
- You do not want to operate a server: you install it, update it and back up its data volume
- Running it yourself does not change where your code goes: it still goes to the model provider you choose
- Status today: Onboarding founding customers
Frequently asked questions
No. It is self-hosted: you install it on a Linux server you operate, with Docker or a one line installer. The only part Reqursor hosts is the license server that activates it.
On the data volume of your own server, encrypted at rest, next to the state, the license, the project workspaces and the run artifacts. That volume is the one thing you need to back up.
The server, its storage and its updates. You pick Docker or a one line installer, back up the data volume, make sure the tools your checks need are installed and confirm updates from Settings, which take a backup first.
An expired license stops new work at the next ticket boundary. It never deletes data or touches your repositories, so your data volume and your repositories are left as they were.
Related pages
Use case
Security sensitive teams
Reqursor Development installs on a Linux server you operate. The agent runs in a sandbox that never sees your data directory, your license or your real API keys, and your code goes only to destinations you configure.
Use case
Control AI spend
Every project has spending caps per ticket, per day and per calendar month. When a cap is hit, Autopilot starts nothing new, Run now is blocked and the message names the cap that stopped it.
Comparison
Versus editor AI assistants
An editor assistant works inside a developer’s session. Reqursor Development takes tickets from a queue, runs on a Linux server you operate and ships only through your checks and evidence.
See it on your own codebase
See Reqursor Development work through real tickets on a repository like yours, in a 30-minute call with the people who build it.