Flutter Surf MCP - Connect to and control Flutter applications programmatically
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"iflow-mcp-erickjtorres-flutter-surf-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# Flutter Surf MCP Flutter Surf MCP is a MCP server that allows interactions with Flutter applications through client LLMs. It provides tools to inspect, control, and automate Flutter applications programmatically. ## Features - Connect to running Flutter applications via VM Service - Retrieve and inspect widget tree state - Click on widgets - Enter text into input fields - Find widgets by text, key, or type - Scroll widgets into view - Perform scrolling actions with customizable parameters - Toggle debug paint features ## Requirements - Python 3.7+ - A running Flutter application with VM service enabled - UV package manager (or pip) - Claude AI with MCP support ## Installation 1. Clone the repository: ```bash git clone https://github.com/yourusername/flutter_surf_mcp.git cd flutter_surf_mcp ``` 2. Install dependencies: ```bash uv pip install -r requirements.txt ``` ## Configuration To use Flutter Surf MCP with Claude, add the following configuration to your Claude AI settings: ```json { "mcpServers": { "flutter_mcp": { "command": "[uv_directory]", "args": [ "--directory", "[path]/flutter_surf_mcp", "run", "flutter_surf_mcp.py" ] } } } ``` **Note: Make sure to enable the flutter driver extension:** ```dart import 'package:flutter_driver/driver_extension.dart'; void main() { enableFlutterDriverExtension(); runApp(const HomeScreen()); } ```
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.