twitter-bookmarks-exporter

Exports all X/Twitter bookmarks into individual Markdown files with metadata.

Extracts all bookmarked tweets from X/Twitter into clean Markdown files for Obsidian, Notion, or any note-taking system.

What it does

1. Parses bookmarks from bookmarks.json (captured from browser Network tab).
2. Handles multiple concatenated API pages in a single file — just paste all responses end-to-end.
3. Converts each bookmark into a Markdown file with YAML frontmatter.
4. Saves files to output/bookmarks/ folder, named sequentially like 0001-title.md.
5. Extracts full content for all tweet types:
   • Regular tweets — full text, t.co URLs expanded to real URLs
   • Note tweets — complete long-form text (no truncation)
   • X Articles — title + preview text + article link
6. Includes quoted tweets as blockquotes, media image/video links, and engagement stats.

Prerequisites

Step 1 — Capture your bookmarks from the browser

X/Twitter does not provide a public bookmarks export API, so you need to capture the responses directly from your browser's Network tab. Follow these steps:

1. Log in to https://x.com in Chrome or Firefox.
2. Open DevTools: press F12 (Windows/Linux) or Cmd+Option+I (Mac).
3. Click the Network tab.
4. Navigate to https://x.com/i/bookmarks.
5. In the Network filter box, type Bookmarks to filter relevant requests.
6. Scroll all the way to the bottom of your bookmarks page. Each scroll triggers a new paginated API request — you must scroll through all your bookmarks to capture every page.
7. In the Network panel you will now see multiple Bookmarks?variables=... requests — one per page.

Step 2 — Save the responses to bookmarks.json

For each Bookmarks request in the Network panel:

1. Right-click the request → Copy → Copy Response.
2. Paste the response into a text editor.

Once you have all responses, save them into a single file named bookmarks.json inside the twitter-bookmarks-exporter/ folder. You can paste the responses one after another — the script handles multiple concatenated JSON objects automatically.

Example with two pages:

text

{ "data": { "bookmark_timeline_v2": { ... } } }
{ "data": { "bookmark_timeline_v2": { ... } } }

No special merging needed — just paste all responses end-to-end and save.

The script handles the current X GraphQL API response format as well as the legacy v1 format.

How to run

bash

python scripts/main.py

This generates Markdown files in output/bookmarks/.

Output format

Each file follows this structure:

markdown

---
title: "Example tweet text here"
url: "https://x.com/username/status/123456789"
author: "@username"
saved_at: "2026-04-14"
type: "tweet"  # tweet | note | article
---

# Example tweet text here

**Author:** Display Name (@username)
**Date:** 2026-04-14
**URL:** https://x.com/username/status/123456789
**Stats:** ❤️ 142 · 🔁 38 · 💬 12 · 🔖 95 · 👁 24000

Full tweet content here, with all t.co links expanded to real URLs.

![image](https://pbs.twimg.com/media/example.jpg)   <!-- if media present -->

> **Quoted tweet** by @otheruser            <!-- if quoting another tweet -->
> https://x.com/otheruser/status/987654321
> Quoted tweet text here.

Example files generated

text

output/bookmarks/
├── 0001-the-thing-that-surprises-me-most-about-llms.md
├── 0002-we-are-still-very-much-in-the-early-innings.md
└── ...

Usage in OpenClaw

Agent prompt: "Export my Twitter bookmarks to Markdown using twitter-bookmarks-exporter."

Expected behavior:

1. Confirm bookmarks.json exists in the skill folder.
2. Run python scripts/main.py.
3. Verify files created in output/bookmarks/.
4. Report completion with file count.

Customization

Edit scripts/main.py to:

• Change output path (OUT_DIR)
• Modify Markdown template (bookmark_to_markdown)
• Add filters (e.g., only tweets from certain authors, by date range)

Troubleshooting

• No bookmarks.json: Manually capture from browser Network tab (see Prerequisites).
• No bookmarks found: Ensure the JSON is the raw API response, not a browser-formatted copy.
• Encoding issues: Ensure the file is saved as UTF-8.
• Empty titles: Script auto-generates slugs from content.

License

MIT License. Free to use, modify, and distribute.