CarrotAI
CarrotAI is a cutting-edge AI agent application that delivers real-time streaming chat via Server-Sent Events (SSE) with built-in Model Control Protocol (MCP) integration. It supports concurrent connections to multiple SSE MCP servers and provides us
What is CarrotAI?
CarrotAI is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to carrotai is a cutting-edge ai agent application that delivers real-time streaming chat via server-sent events (sse) with built-in model control protocol (mcp) integration. it supports concurrent conne...
CarrotAI is a cutting-edge AI agent application that delivers real-time streaming chat via Server-Sent Events (SSE) with built-in Model Control Protocol (MCP) integration. It supports concurrent connections to multiple SSE MCP servers and provides us
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- CarrotAI is a cutting-edge AI agent application that deliver
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx carrotaiConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use CarrotAI
CarrotAI is a full-stack AI agent application that delivers real-time streaming chat through Server-Sent Events and integrates natively with the Model Context Protocol. The backend is built with Python (FastAPI + uv) and the frontend uses Flutter, supporting concurrent connections to multiple SSE-based MCP servers simultaneously. It is designed for teams that want to self-host a private AI chat platform with JWT authentication, multi-language UI, and flexible LLM provider support including DeepSeek and others.
Prerequisites
- Python 3.11+ and uv package manager for the backend
- Flutter SDK 3.x for the frontend (optional if using the web build)
- PostgreSQL database for persistent conversation storage
- One or more LLM API keys (e.g. DeepSeek, OpenRouter, or compatible provider)
- An MCP-compatible SSE server to connect to (CarrotAI itself is the MCP client)
Clone the repository
Clone the CarrotAI repository and navigate into the backend directory to begin setup.
git clone https://github.com/Xingsandesu/CarrotAI.git
cd CarrotAI/backendInstall backend dependencies
Use uv to sync all Python dependencies defined in the project.
uv syncConfigure environment variables
Copy the example env file and fill in your PostgreSQL connection string, JWT secret, and any MCP server SSE endpoints you want to connect.
cp .env.example .env
# Edit .env and set:
# DATABASE_URL=postgresql://user:pass@localhost/carrotai
# SECRET_KEY=your-jwt-secret
# MCP_SERVERS=["http://localhost:10000/sse"]Initialize the database
Run the provided initialization scripts to create tables and seed default configuration.
uv run scripts/init_db.py
uv run scripts/init_config.pyStart the backend server
Launch the FastAPI backend. The Swagger API docs will be available at http://127.0.0.1:8000/docs for testing.
python main.pyConnect MCP servers via configuration
Edit backend/config/mcp_servers.json to add your SSE MCP server endpoints. CarrotAI will connect to all listed servers concurrently.
{
"myMcpService": {
"url": "http://localhost:10000/sse",
"env": {"API_KEY": "your-key"}
}
}CarrotAI Examples
Client configuration
CarrotAI acts as an MCP client. To point it at an SSE MCP server, configure mcp_servers.json in the backend config directory. The backend itself does not expose an MCP endpoint by default.
{
"mcpServers": {
"carrotai-backend": {
"command": "python",
"args": ["main.py"],
"cwd": "/path/to/CarrotAI/backend",
"env": {
"DATABASE_URL": "postgresql://user:pass@localhost/carrotai",
"SECRET_KEY": "your-jwt-secret",
"MCP_SERVERS": "[\"http://localhost:10000/sse\"]"
}
}
}
}Prompts to try
Interact with CarrotAI through its Flutter frontend or via the REST API to use connected MCP tools.
- "Use the filesystem MCP tool to list files in my project directory."
- "Connect to the GitHub MCP server and show me my open pull requests."
- "Enable deep thinking mode and analyze the architecture of this codebase."
- "Upload this PDF and summarize the key findings."
- "Chat with DeepSeek about optimizing my database queries."Troubleshooting CarrotAI
Backend fails to start with a database connection error
Ensure PostgreSQL is running and the DATABASE_URL in .env is correct. Run `uv run scripts/init_db.py` after the database is accessible. Test the connection with psql before starting the backend.
MCP server tools are not appearing in the UI
Check that the SSE MCP server URLs in mcp_servers.json are reachable from the backend. The server must expose an SSE endpoint (not stdio). Restart the backend after editing mcp_servers.json, and check backend logs for connection errors.
Flutter frontend cannot reach the backend
Verify that baseUrl in app_config.dart matches your backend address (default http://127.0.0.1:8000). If running on a device or emulator, use your machine's local network IP instead of 127.0.0.1. Also ensure BACKEND_CORS_ORIGINS includes your frontend origin.
Frequently Asked Questions about CarrotAI
What is CarrotAI?
CarrotAI is a Model Context Protocol (MCP) server that carrotai is a cutting-edge ai agent application that delivers real-time streaming chat via server-sent events (sse) with built-in model control protocol (mcp) integration. it supports concurrent connections to multiple sse mcp servers and provides us It connects AI assistants to external tools and data sources through a standardized interface.
How do I install CarrotAI?
Follow the installation instructions on the CarrotAI GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with CarrotAI?
CarrotAI works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is CarrotAI free to use?
Yes, CarrotAI is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.
CarrotAI Alternatives — Similar Coding Agents Servers
Looking for alternatives to CarrotAI? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up CarrotAI in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use CarrotAI?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.