The credit model of AI app builders, and the bill it hides
Lovable, Bolt, and friends bill by credits for building. The real bill starts when the apps you built start running agents. How to budget the runtime, not just the build.
AI app builders (Lovable, Bolt, v0) sell credits — pay per prompt, ship an app. The credits are the visible spend. The bill that grows after launch is the runtime: the agents your shipped app runs, calling APIs per user action.
The hidden line
A generated app with an agentic feature is spend infrastructure: every user action can trigger tool calls, inference, and third-party APIs. The builder's credits paid for the build; nothing budgets the runtime.
Budget the runtime
Wire the guard into the agents your built apps run: a ceiling per app, an allowlist for the APIs it calls, a velocity limit for loops. The builder gets you to launch; the firewall keeps launch profitable. The same logic applies to any shipped agent →