Asahi – Tweet / Like / DM Delete Script (V1) (Fully Open Source)

This repository contains a fully open-source userscript that helps you bulk delete Tweets, Likes, and Direct Messages from your X (Twitter) account. Everything runs directly in your browser console, no extensions required.

---

✨ Features

• Delete Likes – via like.js from your X archive
• Delete Tweets – via tweets.js or tweet-headers.js
• Delete DMs – via direct_message_* files (1:1 and group conversations)
• Skip Errors Automatically – Already deleted or inaccessible items are ignored
• Date Range Filtering – Delete items within a specific date range, including the end date up to 23:59:59.999
• Progress Save/Resume – Save and resume deletion progress using localStorage
• Anti-Spam Pacing – Random delay ranges and breaks to mimic human behavior and avoid detection

---

📥 Before You Start: Download Your X Archive

1. Go to Settings → Privacy & Security → Your Account → Download an Archive of Your Data.
2. It usually takes 1–2 days to prepare; you’ll receive an email when ready.
3. Download and extract the ZIP. Inside you’ll typically find:
   • data/tweets.js or data/tweet-headers.js
   • data/like.js
   • data/direct_message_headers.js, data/direct_message_group_headers.js, data/direct_messages.js, data/direct_message_groups.js

---

🚀 How to Use (via Console)

1. Log into your X account and go to your profile page (https://x.com/your_screen_name).
2. Right-click anywhere on the page → Inspect → open the Console tab.
3. Open the script file from this repo, copy the entire code.
4. Paste it into the console and press Enter.
5. A small control panel will appear.
6. Choose the archive .js files you want to process (Likes, Tweets, DMs) and set a date range if needed.
7. Click Start Deletion. The script will process the selected files, removing items one by one with anti-spam safeguards.
8. Use the Progress button to resume or reset a previous session.

---

🗂 File Mapping

Content to Delete	Archive Files
Likes	data/like.js
Tweets	data/tweets.js, data/tweet-headers.js
Direct Messages	data/direct_message_headers.js, data/direct_message_group_headers.js, data/direct_messages.js, data/direct_message_groups.js

File names may vary depending on your archive version.

---

📈 Improvements and Security Measures

Implemented Improvements

• Updated Timing Intervals (v0.75.2)

  • Replaced fixed delays with random ranges to enhance anti-spam protection:
    • Likes: 1.8-3 seconds
    • Tweets: 2.5-5 seconds
    • DMs: 4-8 seconds
  • Random jitter (±30%) added to avoid predictable patterns.

• Date Range Enhancement (v0.75.2)

  • The to date is now set to the end of the day (23:59:59.999), ensuring the full day is included in the range.

Newly Added Security Features (v0.75.3 - Proposed)

Based on user feedback, the following critical security measures have been integrated:

1. Pattern Breaking

   • A counter tracks operations, triggering a random break (30-90 seconds) every 15-25 operations to disrupt predictable activity patterns.

2. Error Rate Tracking

   • Monitors failure rates during operations. If the error rate exceeds 20%, the script pauses for 1 hour to avoid triggering anti-spam measures.

3. Human-Like Activity Simulation

   • Introduces a 15% chance of a 3-10 second "reading break" to simulate natural user behavior.

Known Limitations and Future Improvements

• Jitter Calculation: The current jitter is linear; a non-linear distribution could improve natural variation.
• Resume Consistency: Resume mode recalculates ids from files, which may fail if the file changes. Strengthening file hash validation is recommended.
• Performance: Large files may slow down resume operations due to repeated parsing. A temporary cache (e.g., sessionStorage) could help.
• Configurable Delays: Users cannot adjust delay ranges via the UI. Adding min-max inputs would enhance flexibility.

---

⚠️ Disclaimer

This project is provided “as-is.” You are responsible for your account and content. The script is designed only for cleaning up your own archive data, not third-party content. Misuse may lead to account suspension. Use at your own risk.

Security Notes

• Anti-spam measures (global pacing, jitter, pattern breaking, 429 handling) are enabled to reduce detection risk.
• Avoid running the script on multiple tabs simultaneously to prevent IP bans.
• The script uses a dynamic authToken from the page; ensure you are logged into X.

---

💜 Thanks

This project is open-source and free. If you find it useful, following @asahi0x is a nice way to say thanks 💜.