AppFactor
Start with one service. Scale to the estate
Every plan ships the same governed pipeline: agents propose, your engineers sign. Pay for the seats and environments you run, nothing else.
Run discovery and MCP Bridge on a single service and see the pull requests an agent actually produces.
All ten capabilities on the services your team owns, with approval gates per repo and environment.
For organisations running agents across several teams, inside their own network.
On-prem or air-gapped, with the evidence auditors ask for and the controls security teams require.
What does not change between plans
Code, credentials and runtime data stay in your tenancy. There is no plan where that is optional.
Agents open pull requests through your pipeline. The agent that writes a change never approves it.
Each run is attributed to an agent, a model and a cost, visible in the console before the invoice.
Before you pick a plan
What counts as a seat?+
Anyone who can approve a gate. Reviewers, engineers and admins are seats; agents are not.
Which models do the agents use?+
Developer and Team run on our managed inference. Business and Enterprise can bring their own keys or run models inside their own perimeter.
Who owns the code the agents write?+
You do. Every change lands as a pull request in your repository under your licence.
What happens when we go over the plan limits?+
Nothing stops. Overage is metered at the rate shown in the console, and you can set a hard cap per environment.
Can we start on Team and move to Business later?+
Yes. The estate map, gates and history carry over; only the deployment model and the contract change.
Not sure which plan? Start with a scan
Thirty minutes, read-only, scoped to one environment. You keep the estate map either way.
Book a demoAppFactorAI agents that connect, modernize, and maintain your enterprise software. Governed. Self-hosted.