Lotus MCP

Lotus MCP

Enables creation of reusable browser automation skills through demonstration by recording user actions in a browser while narrating, then converting those workflows into executable skills that can be invoked through natural language.

Category
Visit Server

README

Lotus MCP

Create browser automation skills by demonstration. Works with Cursor and Claude Code.

What is Lotus?

Lotus lets you create reusable browser automations ("skills") just by showing what you want to do. Demonstrate a workflow in a browser, narrate what you're doing, and Lotus converts it into an executable skill you can run anytime.


Prerequisites

macOS:

brew install sox

Linux:

sudo apt-get install sox libsox-fmt-all

Windows: Download and install SoX from https://sox.sourceforge.net/


Cursor Setup

1. Install

# Install the MCP
npm install -g github:chrisvin-jabamani/lotus-mcp --ignore-scripts

# Install the Cursor extension
curl -L -o /tmp/lotus.vsix https://github.com/chrisvin-jabamani/lotus-mcp/raw/main/extension/lotus-0.1.0.vsix
cursor --install-extension /tmp/lotus.vsix

2. Configure

Create a file at ~/.cursor/mcp.json (global) or .cursor/mcp.json in your project:

{
  "mcpServers": {
    "lotus": {
      "command": "lotus-mcp-cursor",
      "env": {
        "LOTUS_API_KEY": "your-api-key",
        "LOTUS_URL": "https://www.uselotus.ai"
      }
    }
  }
}

Quick setup (copy-paste into terminal):

mkdir -p ~/.cursor
cat > ~/.cursor/mcp.json << 'EOF'
{
  "mcpServers": {
    "lotus": {
      "command": "lotus-mcp-cursor",
      "env": {
        "LOTUS_API_KEY": "your-api-key",
        "LOTUS_URL": "https://www.uselotus.ai"
      }
    }
  }
}
EOF

Get your API key at uselotus.ai

3. Restart Cursor

Quit and reopen Cursor. Look for the green dot next to "lotus" in Cursor Settings → MCP.

4. Create your first skill

Ask Claude:

"Create a skill to check court availability on Rec.us"

The browser opens inside Cursor. Demonstrate the workflow, then say "I'm done."


Claude Code Setup

1. Install

npm install -g github:chrisvin-jabamani/lotus-mcp --ignore-scripts

2. Configure

claude mcp add lotus -- lotus-mcp-cli

Set environment variables when prompted:

  • LOTUS_API_KEY — Get yours at uselotus.ai
  • LOTUS_URLhttps://www.uselotus.ai

3. Create your first skill

Ask Claude:

"Create a skill to check court availability on Rec.us"

The browser opens in Chrome/Safari. Demonstrate the workflow, then say "I'm done."


How It Works

"Create a skill to book a tennis court"
              ↓
    Browser opens automatically
              ↓
    You demonstrate the workflow
    (narrate what you're doing)
              ↓
         "I'm done"
              ↓
    ✅ Skill created and ready to use

Tools

Tool Description
lotus_start_recording Start recording a workflow
lotus_stop_recording Stop and analyze
lotus_refine_skill Convert to executable skill
lotus_refine_status Check progress
lotus_refine_answer Answer AI questions
lotus_cancel_session Cancel current session

Troubleshooting

Installation fails with ENOENT error?

  • Make sure to use --ignore-scripts flag: npm install -g github:chrisvin-jabamani/lotus-mcp --ignore-scripts

"spawn sox ENOENT" or audio errors?

  • Install SoX: brew install sox (macOS) or apt-get install sox (Linux)

Browser doesn't open in Cursor?

  • Check extension installed: Cmd+Shift+P → "Lotus: Open Browser"
  • Look for "Lotus" in the status bar (bottom right)
  • Run command manually if needed

"Failed to connect to Lotus" or "API returned HTML"?

  • Verify your API key is correct at uselotus.ai
  • Check LOTUS_URL is exactly https://www.uselotus.ai
  • Check MCP server logs: In Cursor, go to Output panel → select "Lotus" from dropdown

MCP not loading in Cursor?

  • Verify ~/.cursor/mcp.json exists and has valid JSON
  • Check for green dot in Cursor Settings → MCP
  • Try restarting Cursor completely

License

MIT

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured