How do I install Xtracticle?

Copy the http config for Cursor, Claude Desktop, Claude Code, or VS Code on this page. It uses the publisher's install, not a PlugMCP proxy.

Does Xtracticle need an API key?

No required environment variable is listed on this record.

What transport does Xtracticle use?

Xtracticle is reached over HTTP.

When was Xtracticle last checked?

Last checked 2026-11-08. Read public X (Twitter) posts, threads and X Articles as clean Markdown. Read-only, no auth.

Xtracticle MCP server

Power your agent

Add this server

The publisher's own URL. Not a PlugMCP endpoint.

Add to Cursor Add in Claude Add to VS Code Run on PlugMCP

Host a copy with our keys and logs.

Server URL

https://xtracticle.com/mcp

The Claude button above opens Claude with the name and URL filled in. You still confirm it there.

shClaude Code

1claude mcp add --transport http com.xtracticle-xtracticle https://xtracticle.com/mcp

Run in a terminal.

jsonClaude Desktop

1{2  "mcpServers": {3    "com.xtracticle-xtracticle": {4      "command": "npx",5      "args": [6        "-y",7        "mcp-remote",8        "https://xtracticle.com/mcp"9      ]10    }11  }12}

claude_desktop_config.json

Remote · HTTPPublisher installStep-by-step: add this in Claude

Remote URL or a command on your machine? How the two installs differ

Run on PlugMCP

Host a copy with our keys and logs.

A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.

Launch

Server facts

Remote

Transport: http
Version: 3.2.0
Last checked

Tools

No tool list yet. Remote servers get one after the first health check.

FAQ

Questions

Live: xtracticle.com — or just change x.com to xtracticle.com in any post URL.

Features

Export formats

Format	What you get
PDF	A4, print-ready; title/author/date/source header, images, page numbers, clickable source link. Rendered page by page so long articles never produce blank PDFs and lines are never cut.
Markdown	Headings, bold/italic/strikethrough, links, nested lists, quotes, code, dividers, images, video thumbnails. Optional YAML front-matter (Obsidian properties, Hugo/Jekyll/Astro).
EPUB 3	Reflowable e-book with embedded images and metadata — Kindle (Send to Kindle), Kobo, Apple Books.
ZIP + images	Markdown + every image stored locally — an offline archive that survives deleted posts.
Text	Clean `.txt`.
Obsidian	Copies the Markdown and opens a new note via `obsidian://new`.

Content

📝 X Articles — full Draft.js conversion: cover, inline images, videos, captions, embedded posts
🧵 Threads from any post — paste the first, a middle or the last post; the whole self-thread is assembled (other people's replies are excluded)
💬 Posts — line breaks, photos, videos and quoted posts
🔗 Flexible input — x.com, twitter.com, mobile, fxtwitter/vxtwitter/fixupx, nitter/xcancel links, /i/web/status/, bare IDs, links inside shared text

Workflow

🔁 URL swap — xtracticle.com/{user}/status/{id} opens a post directly, with rich link previews (Open Graph) when shared
📚 Batch mode — up to 20 links → one ZIP of Markdown files
🔖 Bookmarklet and PWA share target (share from the X app on Android)
🔊 Listen — read aloud with the browser's text-to-speech
📱 iPhone/iPad Shortcut — share a post from the X app straight to Xtracticle (public/shortcuts/)
🤖 MCP server + Markdown API — AI assistants read X posts as Markdown (/mcp, /api/markdown/:id)
📜 History, 🌙 dark mode, 🌍 English, Español, Português, 日本語, 简体中文, العربية, Türkçe (Arabic is right-to-left)
📈 Analytics events — extract, download (format), copy, share, listen, batch_extract, errors — in GA4

Getting Started

git clone https://github.com/ahmetdeveci3112-crypto/Xtracticle.git
cd Xtracticle
npm install
npm run dev

The dev server runs at http://localhost:5173. /api/* is served by the same handler as the Worker (src/shared/api.ts), and every landing page is rendered on the fly — no wrangler needed.

Script	Description
`npm run dev`	Vite dev server with the API
`npm run build`	Production build → `dist/` (all landing pages, 404, sitemap)
`npm run preview:worker`	Build and run the real Worker locally with `wrangler dev` (SSR status pages, 404s, `_headers`)
`npm run lint`	Type-check the app and the Worker

Deployment

Cloudflare Workers + Static Assets, deployed automatically on push to main (build: npm run build, deploy: npx wrangler deploy). No environment variables.

wrangler.json:

run_worker_first — only /api/*, /mcp, /video/* and /*/status/* hit the Worker; everything else is served straight from the CDN
not_found_handling: "404-page" — unknown URLs return a real 404 (no soft-404s)
public/_headers — security headers, Link: </llms.txt>, download headers for the iOS Shortcut

Architecture

Browser ──► Cloudflare
             ├─ Static assets (dist/)          /, /x-article-to-pdf, /tr/, … , 404.html, sitemap.xml
             └─ Worker (src/worker.ts)
                 ├─ /api/thread/:id, /api/tweet/:id ──► src/shared/api.ts ──► FxTwitter API
                 │                                        (edge cache, 5 min)
                 └─ /{user}/status/{id} ──► app shell + per-post OG tags + preloaded JSON (noindex)

src/
├── App.tsx              UI: extraction, exports, batch mode, deep links
├── i18n.ts              UI strings (en, tr, es, pt, ja, zh)
├── lib/export.ts        PDF (jsPDF + html2canvas-pro), EPUB, ZIP (fflate), Obsidian
├── lib/url.ts           Input + path parsing
├── lib/analytics.ts     GA4 event helper
├── shared/fx.ts         FxTwitter client (v2 thread endpoint + v1 fallback), shared by Worker & dev server
├── shared/api.ts        JSON API handler with edge caching
├── shared/convert.ts    Posts / threads / Draft.js articles → Markdown + text + metadata (app + Worker)
├── shared/markdown.ts   GET /api/markdown/:id
├── shared/mcp.ts        Remote MCP server (POST /mcp)
├── shared/text.ts       Title extraction
└── worker.ts            Cloudflare Worker (API + status-page SSR)
site/
├── blocks.ts            Types + localized shared blocks (shortcuts, privacy)
├── pages.ts             English pages (one per search intent) + page registry
├── pages-{es,pt,ja,zh,ar,tr}.ts  The same pages, localized (linked by `group`)
├── curated.ts           Weekly “Best X Articles” pages from site/curated/*.json
├── guides.ts            /guides — informational guides
├── mcp-page.ts          /mcp-server — MCP setup docs
├── render.ts            Head/meta/JSON-LD/content/footer rendering
└── plugin.ts            Vite plugin: emits every page, 404.html and sitemap.xml

Heavy export libraries are dynamically imported, so the initial page only loads React + the app.

SEO & AI Discoverability

Layer	Implementation
Intent pages	`/x-article-to-pdf`, `/x-article-to-markdown`, `/x-thread-to-pdf`, `/x-article-to-epub`, `/save-x-articles-to-obsidian`, `/thread-reader-app-alternative` — each with unique, visible HTML content and FAQ
Languages	Every page in 7 languages (`/`, `/es/…`, `/pt/…`, `/ja/…`, `/zh/…`, `/ar/…`, `/tr/…`) + UI in 7 languages; first-visit language banner, remembered choice redirects (never for crawlers)
Crawlable content	Content is static HTML outside the React root — no JS needed to index it
Structured data	`WebApplication`, `FAQPage` (matches visible FAQ), `BreadcrumbList`
i18n	Reciprocal `hreflang` for every translation group + `x-default`
Share pages	`/{user}/status/{id}` get post-specific OG/Twitter tags; `noindex, follow`
Sitemap / 404	Generated at build; real 404 status for unknown URLs
AI	`/llms.txt`, `/llms-full.txt`, AI crawlers allowed in `robots.txt`

API

GET /api/thread/:id  → { tweets: Tweet[], count, isThread }   # self-thread containing :id
GET /api/tweet/:id   → Tweet
GET /api/markdown/:id → text/markdown  # ?format=markdown|text&thread=0|1&front_matter=0|1
POST /mcp            → MCP (Streamable HTTP, stateless), tool read_x_post

Tweet mirrors the FxTwitter status object (text, author, media, quote, article, …). Responses are edge-cached for 5 minutes — please be gentle.

Weekly “Best X Articles”

/best-x-articles is a hand-reviewed weekly selection built from site/curated/*.json (title, author and a one-sentence summary in our own words — never the article text).

npm run curate -- 2026-W41 https://x.com/user/status/123 https://x.com/other/status/456
# → site/curated/2026-w41.json with title/author/lang filled in; write "intro", "topic" and "summary"
npm run build   # issues with an empty summary are skipped automatically

The newest issue is shown on /best-x-articles (its /best-x-articles/<week> page canonicals to the hub until the next issue); older issues stay at /best-x-articles/<week>.

Contributing

Fork the repo
Create a feature branch (git checkout -b feature/amazing-feature)
Commit your changes
Open a Pull Request

Notes:

Landing page copy lives in site/pages.ts; run npm run build to see the generated HTML
Keep src/shared/* free of DOM/Node-only APIs — it runs in both the Worker and the dev server
Test Worker-only behavior (SSR, 404, headers) with npm run preview:worker

License

MIT

Add this server

Host a copy with our keys and logs.

Server facts

Tools

Questions

How do I install Xtracticle?

Does Xtracticle need an API key?

What transport does Xtracticle use?

When was Xtracticle last checked?

Documentation

Features

Export formats

Content

Workflow

Getting Started

Deployment

Architecture

SEO & AI Discoverability

API

Weekly “Best X Articles”

Contributing

License