TaskMaitre Automation

Automated login and data extraction for playground.enoent.org/taskmaitre using Puppeteer.

Features

• Automated login with access code
• Retrieves all projects without a next task
• Takes screenshots at each step for debugging
• Saves results to JSON file
• Supports headless and visible browser modes

Prerequisites

• Node.js (v14 or higher)
• Chrome or Chromium browser installed

Installation

1. Install dependencies:

npm install

2. Set up your Chrome path (optional - if Chrome is not in the default location):

cp .env.example .env
# Edit .env and set your CHROME_PATH

Usage

Basic usage (headless mode):

node puppeteer-login.js

With visible browser (for debugging):

HEADLESS=false node puppeteer-login.js

With custom Chrome path:

CHROME_PATH=/path/to/chrome node puppeteer-login.js

Configuration

You can modify the configuration directly in puppeteer-login.js:

const CONFIG = {
  siteUrl: 'https://playground.enoent.org/taskmaitre',
  accessCode: 'taskmaster2025',
  chromePath: process.env.CHROME_PATH || '/usr/bin/google-chrome',
  headless: process.env.HEADLESS !== 'false',
  timeout: 30000
};

Output

The script generates the following files:

• initial-page.png - Screenshot of the page before login
• after-login.png - Screenshot after entering access code
• error.png - Screenshot if an error occurs
• access-denied.png - Screenshot if access is denied
• projects-without-next-task.json - JSON file with extracted projects

HTTP Client Alternative

If Puppeteer doesn't work in your environment, there's also an HTTP client-based approach:

node login.js

This uses axios + cheerio to make HTTP requests instead of launching a browser. However, it may not work if the site requires JavaScript execution or has complex authentication.

Troubleshooting

403 Access Denied

If you're getting 403 errors:

• The site might be IP-restricted
• Try running from a different network
• Check if the access code has changed
• Verify the site URL is correct

Chrome/Chromium not found

Set the CHROME_PATH environment variable to point to your Chrome installation:

macOS:

export CHROME_PATH="/Applications/Google Chrome.app/Contents/MacOS/Google Chrome"

Linux:

export CHROME_PATH="/usr/bin/google-chrome"
# or
export CHROME_PATH="/usr/bin/chromium-browser"

Windows (PowerShell):

$env:CHROME_PATH="C:\Program Files\Google\Chrome\Application\chrome.exe"

Page structure is different

The script tries multiple selectors to find projects. If it doesn't find any:

1. Run with HEADLESS=false to see what the page looks like
2. Check the screenshots in the output directory
3. Look at the "Page structure" output in the console
4. Modify the selectors in the getProjects() method

Customization

To customize for different sites or data extraction:

1. Change the target site: Modify CONFIG.siteUrl and CONFIG.accessCode

2. Adjust login selectors: Update the accessCodeSelectors array in the login() method

3. Modify project extraction: Update the projectSelectors array and filtering logic in getProjects()

4. Change filtering criteria: Modify the filter in getProjectsWithoutNextTask() to match your needs

Example Output

=== Projects WITHOUT Next Task (3) ===

1. Project Alpha - Status: Completed - No upcoming tasks

2. Project Beta - Status: On Hold

3. Project Gamma - Status: Archived

License

ISC