Flowise MCP Server

An MCP (Model Context Protocol) server that enables AI agents to interact with Flowise instances. Works with Claude Desktop, VS Code GitHub Copilot, Cursor, and other MCP-compatible clients.

Features

• Flow Management: List, create, update, and delete chatflows and agentflows

• Predictions: Send messages to flows and receive responses

• Flow Analysis: Analyze configurations with improvement suggestions

• Assistants: List and inspect Flowise assistants

• Document Stores: Manage document stores for RAG

• Vector Operations: Upsert vectors and query vector stores

• Chat History: Retrieve and delete conversation history

• Variables & Tools: Manage global variables and tools

Related MCP server: Crew AI MCP Server

Requirements

• Python 3.10+ or uv package manager

• A running Flowise instance with API access

Installation

Option A: Using pip

pip install flowise-mcp

Option B: Using uv (recommended)

# Install uv first if you don't have it
# Windows (PowerShell)
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

Then configure your AI client to use uvx flowise-mcp (see below).

Configuration

Environment Variables

Getting your Flowise API Key

1. Open your Flowise instance

2. Go to Settings > API Keys

3. Create a new API key or copy an existing one

Setup by Client

VS Code (GitHub Copilot)

Requires VS Code 1.99+ with GitHub Copilot.

Add to .vscode/mcp.json in your project or to your User Settings (JSON):

{
  "servers": {
    "flowise": {
      "command": "uvx",
      "args": ["flowise-mcp"],
      "env": {
        "FLOWISE_BASE_URL": "https://your-flowise-instance.com",
        "FLOWISE_API_KEY": "your-api-key"
      }
    }
  }
}

Restart VS Code after adding the configuration.

Claude Desktop

Add to your config file:

• Windows: %APPDATA%\Claude\claude_desktop_config.json

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "flowise": {
      "command": "uvx",
      "args": ["flowise-mcp"],
      "env": {
        "FLOWISE_BASE_URL": "https://your-flowise-instance.com",
        "FLOWISE_API_KEY": "your-api-key"
      }
    }
  }
}

Restart Claude Desktop after adding the configuration.

Available Tools

Flow Management

Assistants

Document Stores & Vectors

Chat & Configuration

Usage Examples

"List all my agentflows"

"Send 'Hello' to flow abc123"

"Analyze flow xyz789 and suggest improvements for better accuracy"

"Create a new chatflow named 'Customer Support Bot'"

"Show me all document stores"

"Query the vector store for 'pricing information'"

"Delete chat history for flow abc123"

"List all assistants"

Development

If you want to contribute or run locally:

git clone https://github.com/JuliDir/flowise-mcp.git
cd flowise-mcp

# Using uv
uv sync --extra dev
uv run pytest

# Or using pip
pip install -e ".[dev]"
pytest

Troubleshooting

License

MIT License - see LICENSE for details.

Author

Julian Di Rocco (@JuliDir)