✉️ Scrape DMs
Export your DM conversations. Scrapes message history with timestamps, sender info, and media attachments.
📋 What It Does
This script provides the following capabilities:
- Automated operation — Runs directly in your browser console on x.com
- Configurable settings — Customize behavior via the CONFIG object
- Real-time progress — Shows live status updates with emoji-coded logs
- Rate limiting — Built-in delays to respect X/Twitter's rate limits
- Data export — Results exported as JSON/CSV for further analysis
Use cases:
- Export your DM conversations. Scrapes message history with timestamps, sender info, and media attachments.
- Automate repetitive scrapers tasks on X/Twitter
- Save time with one-click automation — no API keys needed
- Works in any modern browser (Chrome, Firefox, Edge, Safari)
⚠️ Important Notes
Use responsibly! All automation should respect X/Twitter's Terms of Service. Use conservative settings and include breaks between sessions.
- This script runs in the browser DevTools console — not Node.js
- You must be logged in to x.com for the script to work
- Start with low limits and increase gradually
- Include random delays between actions to appear human
- Don't run multiple automation scripts simultaneously
🌐 Browser Console Usage
Steps:
- Go to
x.com/messages - Open browser console (
F12→ Console tab) - Copy and paste the script from
scripts/scrapeDMs.js - Press Enter to run
/**
* DM Scraper
* Export DM conversations (if accessible)
*
* HOW TO USE:
* 1. Go to x.com/messages
* 2. Click on a conversation you want to export
* 3. Open Developer Console (Ctrl+Shift+J or Cmd+Option+J)
* 4. Paste this script and press Enter
*
* NOTE: Due to Twitter's privacy protections, this can only scrape
* visible messages on screen. Scroll through the conversation first.
*
* by nichxbt - https://github.com/nirholas/XActions
*/
(() => {
const CONFIG = {
MAX_MESSAGES: 1000,
SCROLL_DELAY: 1000,
FORMAT: 'json', // 'json', 'txt'
};
const sleep = (ms) => new Promise(r => setTimeout(r, ms));
const download = (content, filename, type) => {
const blob = new Blob([content], { type });
const url = URL.createObjectURL(blob);
const a = document.createElement('a');
a.href = url;
a.download = filename;
a.click();
URL.revokeObjectURL(url);
};
const extractMessages = () => {
const messages = [];
// Try to find message containers
const messageContainers = document.querySelectorAll('[data-testid="messageEntry"], [data-testid="tweetText"]');
messageContainers.forEach((container, index) => {
const text = container.textContent || '';
if (!text.trim()) return;
// Try to determine if sent or received based on styling/position
const parent = container.closest('[class*="message"]') || container.parentElement;
const isSent = parent?.style?.marginLeft === 'auto' ||
parent?.classList?.toString()?.includes('right') ||
container.closest('[data-testid="messageSent"]') !== null;
// Try to get timestamp
const timeEl = container.closest('[role="listitem"]')?.querySelector('time') ||
container.parentElement?.querySelector('time');
const time = timeEl?.getAttribute('datetime') || timeEl?.textContent || '';
messages.push({
id: index,
text: text.trim(),
type: isSent ? 'sent' : 'received',
time,
});
});
return messages;
};
const getConversationInfo = () => {
// Try to get the conversation header/name
const header = document.querySelector('[data-testid="DM_Conversation_Avatar"]')?.closest('div')?.textContent ||
document.querySelector('h2[role="heading"]')?.textContent ||
'unknown_conversation';
return header.replace(/[^a-z0-9]/gi, '_').slice(0, 30);
};
const run = async () => {
if (!window.location.pathname.includes('/messages')) {
console.error('❌ Please go to x.com/messages and open a conversation first!');
return;
}
console.log('💬 Scraping DM conversation...');
console.log('⚠️ Note: Only visible messages can be scraped.');
console.log('💡 Tip: Scroll through the entire conversation first for best results.\n');
const conversationName = getConversationInfo();
const allMessages = new Map();
let scrolls = 0;
let lastCount = 0;
let noNewCount = 0;
// Try to scroll up to get older messages
const messageContainer = document.querySelector('[data-testid="DM_Conversation"]') ||
document.querySelector('[role="main"]');
while (allMessages.size < CONFIG.MAX_MESSAGES && noNewCount < 5) {
const messages = extractMessages();
const beforeCount = allMessages.size;
messages.forEach(msg => {
const key = `${msg.text.slice(0, 50)}_${msg.time}`;
if (!allMessages.has(key)) {
allMessages.set(key, msg);
}
});
const added = allMessages.size - beforeCount;
if (added > 0) {
console.log(`💬 Collected ${allMessages.size} messages...`);
noNewCount = 0;
} else {
noNewCount++;
}
// Scroll up within the message container
if (messageContainer) {
messageContainer.scrollTop -= 500;
}
await sleep(CONFIG.SCROLL_DELAY);
scrolls++;
if (scrolls > 50) break;
}
const messageList = Array.from(allMessages.values());
// Sort by order they appeared (approximate chronological)
messageList.sort((a, b) => a.id - b.id);
console.log('\n' + '='.repeat(60));
console.log(`💬 SCRAPED ${messageList.length} MESSAGES`);
console.log('='.repeat(60) + '\n');
const sent = messageList.filter(m => m.type === 'sent').length;
const received = messageList.filter(m => m.type === 'received').length;
console.log(`📊 Stats: ${sent} sent, ${received} received`);
// Preview
messageList.slice(0, 5).forEach((m, i) => {
const prefix = m.type === 'sent' ? '→' : '←';
console.log(`${prefix} "${m.text.slice(0, 60)}..."`);
});
if (CONFIG.FORMAT === 'json') {
const data = {
conversation: conversationName,
exportedAt: new Date().toISOString(),
messageCount: messageList.length,
messages: messageList,
};
download(JSON.stringify(data, null, 2), `dm_${conversationName}_${Date.now()}.json`, 'application/json');
console.log('💾 Downloaded dm_conversation.json');
}
if (CONFIG.FORMAT === 'txt') {
const txt = messageList.map(m => {
const prefix = m.type === 'sent' ? '[You]' : '[Them]';
return `${prefix} ${m.text}`;
}).join('\n\n');
download(txt, `dm_${conversationName}_${Date.now()}.txt`, 'text/plain');
console.log('💾 Downloaded dm_conversation.txt');
}
window.scrapedDMs = { conversation: conversationName, messages: messageList };
console.log('\n✅ Done! Access data: window.scrapedDMs');
console.log('\n⚠️ Privacy note: Be mindful of sharing DM content.');
};
run();
})();
⚙️ Configuration
| Setting | Default | Description |
|---|---|---|
MAX_MESSAGES |
1000 |
M a x m e s s a g e s |
SCROLL_DELAY |
1000 |
S c r o l l d e l a y |
FORMAT |
'json', |
'json', 'txt' |
📖 Step-by-Step Tutorial
Step 1: Navigate to the right page
Open your browser and go to x.com/messages. Make sure you're logged in to your X/Twitter account.
Step 2: Open the browser console
- Chrome/Edge: Press
F12orCtrl+Shift+J(Mac:Cmd+Option+J) - Firefox: Press
F12orCtrl+Shift+K - Safari: Enable Developer menu in Preferences → Advanced, then press
Cmd+Option+C
Step 3: Paste the script
Copy the entire script from scripts/scrapeDMs.js and paste it into the console.
Step 4: Customize the CONFIG (optional)
Before running, you can modify the CONFIG object at the top of the script to adjust behavior:
const CONFIG = {
// Edit these values before running
// See Configuration table above for all options
};
Step 5: Run and monitor
Press Enter to run the script. Watch the console for real-time progress logs:
- ✅ Green messages = success
- 🔄 Blue messages = in progress
- ⚠️ Yellow messages = warnings
- ❌ Red messages = errors
Step 6: Export results
Most scripts automatically download results as JSON/CSV when complete. Check your Downloads folder.
🖥️ CLI Usage
You can also run this via the XActions CLI:
# Install XActions globally
npm install -g xactions
# Run via CLI
xactions --help
🤖 MCP Server Usage
Use with AI agents (Claude, Cursor, etc.) via the MCP server:
# Start MCP server
npm run mcp
See the MCP Setup Guide for integration with Claude Desktop, Cursor, and other AI tools.
📁 Source Files
| File | Description |
|---|---|
scripts/scrapeDMs.js |
Main script |
🔗 Related Scripts
| Script | Description |
|---|---|
| Scrape Profile with Replies | Scrape a profile's tweets AND replies |
| Scrape Analytics | Scrape your account and post analytics |
| Scrape Bookmarks | Scrape all your bookmarked tweets |
| Scrape Cashtag Search | Scrape cashtag search results with sentiment analysis |
| Scrape Explore | Scrape the Explore page trends and content |
Author: nich (@nichxbt) — XActions on GitHub
⚡ Ready to try Scrape DMs?
XActions is 100% free and open-source. No API keys, no fees, no signup.
Browse All Scripts