tweet.md logotweet.md

X/Twitter Markdown comparison

tweet.md vs Jina Reader on a quoted X post

We ran the same Sam Altman post through both tools. tweet.md keeps the quoted OpenAI post as a separate, sourced block. Jina Reader flattens it into the surrounding x.com page and adds replies, login prompts, and navigation text.

This is an X-specific verdict. Jina Reader handles general web pages. tweet.md is built for X posts, threads, Articles, and profiles.

Result

Jina captures the page. tweet.md captures the post.

Both outputs contain Sam Altman's sentence and the OpenAI report text. The difference is whether an agent can tell where the quote starts, who wrote it, and which engagement numbers belong to it.

What is preservedtweet.mdJina Reader
Quoted postA separate Quoted post block with its own author, ID, source, time, stats, and text.Flattened into the scraped page text, with nothing marking where the quote starts or who wrote it.
Unrelated repliesExcluded from the single-post response.Included after the post together with reply authors, engagement, and images.
Page chromeNo login prompts, trending links, or navigation text.Includes login, sign-up, relevant people, trending, and footer content from x.com.
X metadataPost IDs, source URLs, timestamps, author details, engagement, and thread metadata.Page title and URL, plus whatever details happen to be visible in the rendered page.
Best fitX posts, threads, Articles, and profiles for agents, archives, and research.General web pages that need a broad URL-to-Markdown reader.

Actual output

The same URL, two different documents

These excerpts come from outputs captured on 2026-08-27. Long reply text and repeated page chrome are shortened, but the quote structure is shown as returned.

Jina Reader

Open output

A broad scrape of the rendered x.com page. The quote appears inline, then the output continues into replies and account prompts.

Title: Sam Altman (@sama) on X
URL Source: https://x.com/sama/status/2092712656096358527

## Post
[Log in](...) [Sign up](...)

Sam Altman @sama
> this is a good report about a bad thing:

OpenAI @OpenAI
> We have conducted a thorough investigation into the Hugging Face incident.
> We are releasing a technical report and accompanying blog post...

Replies included in the page stream:
- Jimmy Apples: RSI wen ?
- Yuli Brown: good report about a bad thing...

## Log in or sign up for X
## Relevant people
## Trending now

tweet.md

Open output

An X-aware document. The main post and quoted post remain separate records with their own source and metadata.

# 1/1 — Post by Sam Altman (@sama)

Post ID: 2092712656096358527
Source: https://x.com/sama/status/2092712656096358527
Posted: 2026-08-26T20:35:52.000Z
Reply to: none

Text:

> this is a good report about a bad thing: https://twitter.com/OpenAI/status/2092691861773160673

Quoted post:

Author: OpenAI (@OpenAI)
Post ID: 2092691861773160673
Source: https://x.com/OpenAI/status/2092691861773160673
Posted: 2026-08-26T19:13:14.000Z
Reply to: none

Stats:
- Replies: 618
- Reposts: 1,231
- Quotes: 847
- Likes: 9,955
- Bookmarks: 9,142
- Impressions: 9,211,297

Text:

> We have conducted a thorough investigation into the Hugging Face incident.
>
> We are releasing a technical report and accompanying blog post that reconstruct the agents’ activity, explain why existing safeguards failed, and detail how we’re preventing recurrence.
>
> https://openai.com/index/hugging-face-incident-and-the-road-ahead/

# Thread Metadata

Posts: 1
Root author: Sam Altman (@sama)

Which one should you use?

Pick the reader that understands the source

Use tweet.md
When an agent, archive, or research workflow needs X-specific relationships, authors, stats, thread order, Articles, or profiles.
Use Jina Reader
When you need one general reader for many websites and source-specific X structure is not required.

For X/Twitter posts, tweet.md produces the cleaner document because it reads the post as data instead of treating it as another web page. Read the API documentation for thread scopes and output controls.