Make a LinkedIn carousel from Claude, Cursor or ChatGPT with the slidler MCP

Slidler runs a free remote MCP server at https://slidler.com/mcp. There's no sign-up and no API key. Add it to your AI assistant, then ask it to turn an article link, a YouTube video or your own text into a carousel. The assistant calls slidler's create_carousel tool and gets back the slide text (a cover, 6–8 key points and a closing slide) plus a link to a 1080×1350 PDF you can upload to LinkedIn as a document post.

The MCP URL

https://slidler.com/mcp

Remote MCP over Streamable HTTP (JSON responses). No auth, no key, no sign-up.

What it does

Reads a public article or blog URL, a YouTube video (from its captions, or its description when there are none) or pasted text.
Builds a carousel from the source's own sentences: a cover, 6–8 key points and a closing call to action.
Renders a 1080×1350 (4:5) vector PDF, the size that fills the LinkedIn mobile feed.

Install

Claude (claude.ai and Claude Desktop)

Open Settings → Connectors → Add custom connector. Name it Slidler, paste the URL below and save. No OAuth or key is needed. Custom connectors depend on your Claude plan.

https://slidler.com/mcp

Cursor

Add this to ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (one project):

{
  "mcpServers": {
    "slidler": { "url": "https://slidler.com/mcp" }
  }
}

VS Code (GitHub Copilot agent mode)

Add this to .vscode/mcp.json:

{
  "servers": {
    "slidler": { "type": "http", "url": "https://slidler.com/mcp" }
  }
}

Claude Code

Run in your terminal:

claude mcp add --transport http slidler https://slidler.com/mcp

ChatGPT

In ChatGPT plans that allow custom connectors, turn on developer mode in Settings (under Apps/Connectors, Advanced), create a custom connector with the URL below and choose no authentication.

https://slidler.com/mcp

Other MCP clients

Any client that supports remote Streamable HTTP servers works with the URL. For stdio-only clients, use a bridge such as npx mcp-remote https://slidler.com/mcp.

Example prompt

Use slidler to turn https://paulgraham.com/ds.html into a LinkedIn carousel with 6 points. Show me the slides and the PDF link.

Other ideas: “Make a carousel from this YouTube video: …”, or paste your draft post and say “Turn this into a slidler carousel with 8 points.”

What you get back

The slide text as JSON (title, cover, points, closing slide), so the assistant can show or edit it.
A link to the PDF (it expires after about 15 minutes, so download it right away).
Every page carries a small “made with slidler.com” line.

Want to edit the design, pick one of 6 templates or export PNGs for Instagram? Use the web maker. It's free and needs no sign-up too.

Where slidler is listed

Developers: see llms.txt and the OpenAPI contract for the REST API.

FAQ

Is the slidler MCP server free?

Yes. No account and no API key. It's rate-limited per IP, shared with the public API.

Which tools does it expose?

One tool, create_carousel. It takes source (a public URL or at least 40 characters of text), an optional title (for text only) and an optional n (6 to 8 points).

Does it write new content with AI?

No. Slidler picks the strongest sentences from your source, one per slide, so nothing is invented. Your assistant can rewrite the text and send it back as source.

How long does the PDF link work?

About 15 minutes. Download the PDF soon after the assistant shows the link, or ask it to create the carousel again.

Is there a watermark?

Yes. Every page carries a small “made with slidler.com” line.

Can it post to LinkedIn for me?

No. It returns the PDF; you upload it with “Add a document” in a new LinkedIn post.

More: ChatGPT to LinkedIn carousel · How to post a carousel on LinkedIn · LinkedIn carousel size · AI carousel maker