Real Agent Cases

Turn X bookmarks and likes into a searchable knowledge base

Posted
by Sheeki (@sheeki03)
Last checked
Method:
How we verify

Evidence: Described by posterThe person describes what happened in words, without screenshots or video.

Outcome

The poster scraped all their X bookmarks and likes with Claude in Chrome and other tools, cleaned the linked pages with Firecrawl, and built a RAG chatbot over everything they had saved.

Agent

Claude in Chrome, Vercel Agent Browser

Prompt Recreated prompt

The source didn't share its exact prompt, so we wrote this from what it describes.

Go through my X bookmarks and likes and export every post with its text, author, date and any external links into a JSON file.

Try this in:

ChatGPT Gemini Grok

Sends the Recreated prompt shown above. ChatGPT and Grok open with it filled in, and ChatGPT may send it straight away. Gemini opens empty, so we copy the prompt for you to paste. Every button also copies it.

The source used Claude in Chrome and Vercel Agent Browser, which can take actions in a browser or app. These are general chat tools, so results may differ, and a normal chat may only give you a plan.

Opens in a new tab. We don't see what you send.

These links are only a shortcut. We're not affiliated with OpenAI, Google or xAI, and your results may not match the source.

Optional stronger version

Our rewrite with extra checks. The poster didn't test it.

Goal: Export all my X bookmarks and likes to a JSON file with text, author, date and external links.

Inputs (fill in or confirm):
- My X account (I'll log in)
- Output file path
- Whether to include media links and quoted posts

Rules:
- Use the official export or API where possible; scroll-scraping only if needed and slowly.
- For each post capture id, URL, author handle, date (ISO), full text, external links, and source (bookmark or like).
- De-duplicate posts that are both liked and bookmarked.
- Save progress every 200 posts so a stop doesn't lose work.
- Scroll slowly to avoid rate limits; pause if X shows errors.
- Save progress to the file in batches so nothing is lost.
- Record post URL, author, date, text and linked URLs for each item.
- Don't like, unlike or remove bookmarks.

If you get stuck: If X rate-limits you, wait and resume from the last saved post rather than starting over. If a login, CAPTCHA or two-step code appears, stop and ask me to handle it; never guess or store passwords.

Verify before you finish: Compare the count exported with the count X shows, and validate the JSON.

Output: The JSON file, a count summary and any posts that failed.

Done when: the JSON loads cleanly and the counts match.

Source

https://x.com/sheeki03/status/2013898161782481048

Related cases

  • Make years of calls, podcasts and posts searchable with Claude CodeRecreated promptEvidence: Described by poster

    Claude Code organized years of calls, podcasts, tweets and presentations into a searchable database connected to AI, with a daily job that adds new content; asked for a masterclass outline, it drew on 224 chunks from 78 sources.

    Claude Code · Content · Source date:

  • Set up a Google Analytics dashboard with Claude in ChromeImproved promptEvidence: Described by posterFailure

    Claude in Chrome spent 45 minutes building a Google Analytics dashboard, then couldn't work out how to drag one element, so it undid all of its work and told the poster it couldn't do it.

    Claude in Chrome · Data · Source date:

  • Sort a Google Sheet in a custom priority order with Claude in ChromeImproved promptEvidence: Screenshots or logsFailure

    Asked to sort a Google Sheet in a custom order (critical, high, medium, low), Claude for Chrome kept trying to sort it alphabetically instead, per the poster's screenshot.

    Claude in Chrome · Data · Source date: