Skip to content
MCP server

SnapClap MCP

Turn public social video links into timestamped PDFs on your own machine

Runs locally
Install
Platforms
CLI
Runs
Locally
Version
0.1.0
Listing updated
4 days ago

Similar MCP servers

About SnapClap MCP

SnapClap MCP is an MCP server for design and media. It runs locally on your machine and is started by your AI client.

Processing happens locally. The MCP server downloads an accessible video straight from its source, extracts frames with FFmpeg on your computer and writes the PDF there. The developer page states SnapClap runs no service that receives or stores video or PDF files, and that no video URLs or conversion analytics are sent to it. Temporary source videos are deleted after conversion, while frames, the PDF and job metadata stay in a local job folder.

Each PDF has an overview linked to timestamped frame pages, intended as a reusable visual reference for notes or AI conversations. Requirements are Node.js 20.19 or later and FFmpeg on the PATH; X and Instagram fall back to yt-dlp when needed, and Frame.io public shares need Chromium. Only public links work and sign-in walls are not bypassed.

Setup is one command such as claude mcp add snapclap -- npx -y @calnflow/snapclap-mcp, with equivalents for Codex and Cursor. The tool names are not listed in the available text.

Install

Run in a terminal

claude mcp add snapclap-mcp -- npx -y @calnflow/snapclap-mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "snapclap-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@calnflow/snapclap-mcp"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "snapclap-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@calnflow/snapclap-mcp"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "snapclap-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@calnflow/snapclap-mcp"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "snapclap-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@calnflow/snapclap-mcp"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "snapclap-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@calnflow/snapclap-mcp"
      ]
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

What it does

Local conversion
Download, frame extraction and PDF creation run on your computer.
Timestamped PDF
Overview page links to frame pages with timestamps.
Public links only
Does not bypass private posts or sign-in walls.
Frame.io shares
Supports public Frame.io share links.
Local job folder
Keeps frames, PDF and metadata where the agent can reopen them.
Easy install
Runs via npx for Claude Code, Codex and Cursor.

Good for

  • Turn a podcast clip into a PDF reference
  • Give an agent a visual summary of a video
  • Save key moments of a talk with timestamps
  • Archive footage frames for project notes
  • Review a Frame.io share as a PDF

SnapClap MCP FAQ

Is my video sent to SnapClap?

No. The page says processing is local and no video or PDF files are received or stored by SnapClap.

What do I need installed?

Node.js 20.19 or later and FFmpeg on PATH; Chromium for Frame.io shares.

Does it work on private videos?

No. Only public links are supported.

Which clients are documented?

Codex, Cursor and Claude Code, plus Claude Desktop with the same server entry.

How do I add SnapClap MCP to Claude Code?

Run this command in your terminal: claude mcp add snapclap-mcp -- npx -y @calnflow/snapclap-mcp

What are the best alternatives to SnapClap MCP?

The closest alternatives to SnapClap MCP are Katto MCP, ClipMind, Mochify, Video to Audio Converter MCP and Clipspeed MCP.

Where this comes from

  • Description: written from the project's README and the project's website.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers