For AI · Claude

Convert X (Twitter) to Markdown for Claude

Paste any X article, thread, or tweet. Get clean, prompt-ready Markdown that Claude can actually read — no HTML, no scraping, no rate-limit dance. Drop the output straight into a chat, a Project, or Claude Code.

Why feed X content through xmd before Claude

Claude is excellent at reading prose. It is less good at reading the JavaScript-rendered soup that x.com serves to anonymous clients. When you paste a tweet URL into a Claude chat, the model fetches the page and often gets back boilerplate, a login wall, or a fragment of the post body — not the actual content you wanted it to reason about.

xmd resolves the post server-side, normalises threads into ordered Markdown sections, preserves quoted-tweet context, and strips everything Claude does not need. The output is a single .md document that reads top-to-bottom the way the original post was meant to be read.

Two ways to use xmd with Claude today

1. Paste into a Claude chat

Paste an X URL into the xmd converter. Copy the Markdown. Paste it into a Claude conversation as context for whatever you want Claude to do next — summarise, fact-check, extract entities, write a counter-argument, translate.

2. Add to a Claude Project as a knowledge document

Save the Markdown as a .md file and drop it into a Claude Project. Threads collapse into a single document with stable heading anchors, which keeps Project search and inline citations meaningful — Claude can cite a specific section of the thread, not just the URL.

Programmatic access for agents

An HTTP API for Claude agents and Claude Code workflows — converting URLs in batch, streaming Markdown into a system prompt, processing a queue of X links — is on the waitlist. The web converter is the only shipped surface today.

What the output looks like

xmd returns a single Markdown document with the post's title, author handle, and timestamp as front-matter, the body as ordered sections, and quoted tweets as nested blockquotes. No tracking parameters, no script tags, no HTML entities. Claude treats it as plain prose — which is the whole point.

Frequently asked questions

How do I get an X (Twitter) post into Claude?

Paste the X URL into the xmd converter, copy the Markdown output, and paste it into Claude. The output strips HTML, normalises threads into ordered sections, and keeps quoted-tweet context — so Claude reads the post the way a human would.

Can I add a converted post to a Claude Project?

Yes. xmd outputs a clean .md file you can drop into any Claude Project as a knowledge document. Threads collapse into a single Markdown file with stable headings, which keeps Project search and citations meaningful.

Why not just paste the X URL directly into Claude?

Claude can fetch URLs, but X aggressively rate-limits unauthenticated requests and renders most content via JavaScript. The fetched HTML often arrives stripped of the actual post body. xmd resolves the post server-side and returns the text Claude needs to reason about.

Is there a programmatic API for Claude agents?

Not publicly yet. The web converter is shipped and free; an HTTP API for production Claude agents is on the waitlist — join from this page to get notified when it ships.

Is xmd free to use with Claude?

Yes. The web converter is free and requires no login.


Ready to try it? Open the xmd converter and paste an X URL. The output is in your clipboard in under 200ms.