LCBro

Low Cost Browsing MCP Server - MCP server aggregator for browser automation, parsing and optional LLM data cleaning using Playwright.
  • typescript

5

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "lcbro-lcbro-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "lcbro"
      ],
      "env": {
        "LLM_HOST": "localhost",
        "LLM_PORT": "11434",
        "OPENAI_API_KEY": "sk-...",
        "ANTHROPIC_API_KEY": "...",
        "LLM_DEFAULT_MODEL": "ollama:llama3.1"
      }
    }
  }
}

LCBro MCP Server enables browser automation via Playwright and CDP with intelligent preprocessing, robust logging, and remote browser management. It focuses on cost-efficient data preparation and flexible deployment, so you can automate complex web flows while keeping LLM usage lean and responsive.

How to use

You use LCBro by running the MCP server locally or remotely and connecting your MCP client to it. The server exposes browser automation capabilities, logging, and preprocessing features that help you control browser sessions, perform interactions, extract data, and manage contexts. Start with a standard configuration, then tailor CDP support, logging, and preprocessing to fit your workflow. The MCP client sends high-level commands like navigate, interact, and extract, while LCBro handles the underlying browser automation, data cleaning, and LLM-assisted preprocessing.

How to install

Prerequisites: You need Node.js 18+ and npm or yarn, plus a Chromium-based browser for CDP support.

# Install LCBro globally (recommended for quick start)
npm install -g lcbro

# Verify installation
lcbro --version

# Start the server with default config
lcbro

# Start with CDP support
lcbro --cdp-enabled

Configuration and usage notes

LCBro uses YAML configuration with CLI overrides. You can enable CDP, customize ports, and adjust LLm preprocessing settings to balance cost and latency. Environment variables allow you to switch between local LLMs (Ollama, JAN) or external APIs.

server:
  host: "localhost"
  port: 3000

browser:
  engine: playwright  # options: playwright | cdp
  headless: true
  defaultTimeoutMs: 30000
  maxContexts: 8
  storageDir: /data/profiles

llm:
  defaultModel: "ollama:llama3.1"
  maxOutputTokens: 2000
  temperature: 0
  host: "localhost"
  port: 11434
  janPort: 1337
  autoPreprocess: true

security:
  allowDomains: ["example.com", "gov.br"]
  blockPrivateNetworks: true

You can override these settings from the command line, for example enabling CDP support or setting a custom log directory.

## MCP client integration example

If you are wiring LCBro into an MCP client, you can use a standard stdio configuration that runs LCBro as a local process.

{ "mcpServers": { "lcbro": { "command": "npx", "args": ["-y", "lcbro"] } } }


## Available tools

### navigate\_open

Open a new URL in a browser context using the configured browser and context.

### navigate\_goto

Navigate to a URL in an existing browser context.

### interact\_click

Click on elements by selector, text, or role to simulate user interactions.

### interact\_type

Type text into input fields or editable elements.

### interact\_wait

Wait for a condition such as a selector, network idle, or URL change.

### extract\_content

Extract page content in various formats for downstream processing.

### extract\_table

Extract table data and return it as JSON.

### extract\_attributes

Extract attributes from selected elements.

### extract\_screenshot

Take screenshots of pages or specific elements.

### session\_auth

Perform authentication sequences within a browser session.

### session\_manage

Manage browser sessions and contexts.

### llm\_transform

Transform data using LLM with custom instructions.

### llm\_analyze

Analyze content with LLM preprocessing.
Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
LCBro MCP Server - lcbro/lcbro-mcp | VeilStrat