Lotus

Lotus MCP lets you create reusable browser automation skills by demonstration. You narrate a browser workflow, and Lotus converts it into an executable skill you can run on demand. This guide walks you through using the Lotus MCP server with a client, installing prerequisites, configuring the setup, and creating your first skills.

How to use

You use an MCP client to connect to the Lotus server and create skills by demonstrating browser workflows. Start the MCP client, ensure your Lotus configuration is active, and then instruct the client to record a workflow. Demonstrate the steps you want the skill to perform in the browser, narrate what you’re doing, and finish with a clear indication like "I’m done". The client will convert your demonstration into an executable skill that you can run later.

How to install

Prerequisites for macOS, Linux, and Windows are provided to set up the environment and required tools.

brew install sox

Install SoX on Linux.

sudo apt-get install sox libsox-fmt-all

Windows users should download and install SoX from the official site.

https://sox.sourceforge.net/

Cursor MCP setup

Install the MCP and the Cursor extension, then configure the MCP server for Cursor in your user or project settings.

# 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

Create the MCP configuration for Cursor. Place it 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"
      }
    }
  }
}

Alternatively, copy the block into a terminal to create the file.

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

Restart Cursor to apply changes. Look for the green dot next to “lotus” under Cursor Settings → MCP.

Create your first skill by asking Claude to 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

Install the Claude Code MCP client, then configure the Claude integration to access Lotus.

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

Add Lotus as an MCP with Claude. This connects Claude’s MCP workflow with Lotus.

claude mcp add lotus -- lotus-mcp-cli

Create your first Claude-driven skill

Set environment variables when prompted:

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

How it works

The flow starts with your instruction to create a skill, the browser opens, you demonstrate the workflow, and when you say you’re done, the skill is created and ready to use.

Tools

The following commands help you manage and control the Lotus MCP workflow within your session.

Troubleshooting

If installation fails with ENOENT, ensure you use the --ignore-scripts flag when installing the MCP.

For audio-related errors, install SoX on your platform.

If the browser doesn’t open in Cursor, verify the extension is installed and try the in-app command to open the browser.

To troubleshoot connection issues, verify your API key and URL, and check MCP logs in Cursor’s Output panel.

If the MCP does not load, confirm the MCP JSON file exists and the green dot is visible in Cursor Settings → MCP, then restart Cursor.

License

MIT

Available tools

lotus_start_recording

Start recording a browser workflow to convert into a skill.

lotus_stop_recording

Stop recording and analyze the captured workflow.

lotus_refine_skill

Convert the recorded workflow into an executable skill.

lotus_refine_status

Check the progress of skill creation and refinement.

lotus_refine_answer

Answer AI questions during the skill development process.

lotus_cancel_session

Cancel the current recording/refinement session.