The most requested feature in FlutterFlow history has finally landed.
You can now bring AI agents directly into FlutterFlow and switch between them. Claude Code, Gemini CLI, Codex, or any MCP-compatible client. Your agent works within your project with full context for your widgets, logic, and data sources, not as an external tool you copy-paste results into.
Why it matters
Until now, using AI with FlutterFlow meant working across two surfaces. You'd prompt an agent in one window and manually apply the results in another. That friction adds up fast, especially on complex projects.
MCP closes that gap. Your agent is in the loop, understands your app's structure, and acts on it directly. The result is faster iteration, more accurate output, and a workflow that feels native rather than bolted on.
Your agent, your choice (BYOA)
Because MCP is an open standard, you're never locked into a single provider. Switch between Claude Code, Gemini CLI, Codex, or any other MCP-compatible client. Whatever fits your team's workflow.
Check this link to find out how to get started pub.dev/packages/flutterflow_cli.