Skip to main content
Glama
afshawnlotfi

Configurable Puppeteer MCP Server

by afshawnlotfi
README.md
# Configurable Puppeteer MCP Server

A Model Context Protocol server that provides browser automation capabilities using Puppeteer with configurable options. This server enables LLMs to interact with web pages, take screenshots, and execute JavaScript in a real browser environment, with the ability to customize Puppeteer launch options through environment variables.

## Components

### Tools

- **puppeteer_navigate**
  - Navigate to any URL in the browser
  - Input: `url` (string)

- **puppeteer_screenshot**
  - Capture screenshots of the entire page or specific elements
  - Inputs:
    - `name` (string, required): Name for the screenshot
    - `selector` (string, optional): CSS selector for element to screenshot
    - `width` (number, optional, default: 800): Screenshot width
    - `height` (number, optional, default: 600): Screenshot height

- **puppeteer_click**
  - Click elements on the page
  - Input: `selector` (string): CSS selector for element to click

- **puppeteer_hover**
  - Hover elements on the page
  - Input: `selector` (string): CSS selector for element to hover

- **puppeteer_fill**
  - Fill out input fields
  - Inputs:
    - `selector` (string): CSS selector for input field
    - `value` (string): Value to fill

- **puppeteer_select**
  - Select an element with SELECT tag
  - Inputs:
    - `selector` (string): CSS selector for element to select
    - `value` (string): Value to select

- **puppeteer_evaluate**
  - Execute JavaScript in the browser console
  - Input: `script` (string): JavaScript code to execute

### Resources

The server provides access to two types of resources:

1. **Console Logs** (`console://logs`)
   - Browser console output in text format
   - Includes all console messages from the browser

2. **Screenshots** (`screenshot://<name>`)
   - PNG images of captured screenshots
   - Accessible via the screenshot name specified during capture

## Key Features

- Browser automation
- Console log monitoring
- Screenshot capabilities
- JavaScript execution
- Basic web interaction (navigation, clicking, form filling)
- **Configurable Puppeteer options** through environment variables

## Configuration

### Using with Custom Puppeteer Options

You can configure Puppeteer launch options by providing a JSON string in the `PUPPETEER_ARGS` environment variable. This allows you to customize browser behavior without modifying the server code.

#### Example: Using Firefox Instead of Chrome

```json
{
  "mcpServers": {
    "puppeteer": {
      "command": "npx",
      "args": ["-y", "github:afshawnlotfi/mcp-configurable-puppeteer"],
      "env": {
        "PUPPETEER_ARGS": "{\"browser\": \"firefox\"}"
      }
    }
  }
}
```

#### Example: Configuring Browser Window Size

```json
{
  "mcpServers": {
    "puppeteer": {
      "command": "npx",
      "args": ["-y", "github:afshawnlotfi/mcp-configurable-puppeteer"],
      "env": {
        "PUPPETEER_ARGS": "{\"defaultViewport\": {\"width\": 1280, \"height\": 800}}"
      }
    }
  }
}
```

### Standard Configuration

#### NPX
```json
{
  "mcpServers": {
    "puppeteer": {
      "command": "npx",
      "args": ["-y", "github:afshawnlotfi/mcp-configurable-puppeteer"]
    }
  }
}
```

You can also specify a branch, tag, or commit:

```json
{
  "mcpServers": {
    "puppeteer": {
      "command": "npx",
      "args": ["-y", "github:afshawnlotfi/mcp-configurable-puppeteer#main"]
    }
  }
}
```


## License

This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License. For more details, please see the LICENSE file in the project repository.

TDQS

B3.3/5.0

Scored across 7 tools

Disambiguation5/5

Every tool has a clearly distinct purpose with no ambiguity. Clicking, evaluating JavaScript, filling inputs, hovering, navigating, taking screenshots, and selecting dropdown elements are all separate browser automation operations that an agent can easily distinguish. The descriptions clearly differentiate each action.

Naming Consistency5/5

All tools follow a perfect verb_noun pattern with consistent snake_case naming (puppeteer_click, puppeteer_evaluate, etc.). The 'puppeteer_' prefix creates a clear namespace, and the action verbs (click, evaluate, fill, hover, navigate, screenshot, select) are consistently descriptive and parallel in structure.

Tool Count5/5

Seven tools is well-scoped for a Puppeteer automation server. This provides core browser interaction capabilities without being overwhelming. Each tool earns its place by covering essential operations like navigation, interaction, evaluation, and capture that form a complete automation workflow.

Completeness4/5

The toolset covers most essential Puppeteer operations well, including navigation, interaction, evaluation, and screenshots. Minor gaps exist - there's no explicit tool for waiting/loading states, keyboard input, or cookie/session management - but agents can work around these using evaluate or other tools. Core browser automation workflows are well-supported.

Maintenance

ActivityInactive
ResponsivenessNo issues