MCP Integration

WiseScribe exposes a Model Context Protocol (MCP) server at https://wisescribe.ai/api/mcp. Connect it to Claude Desktop, Cursor, Zed, or any MCP-compatible tool and let your AI assistant transcribe videos directly from your chat.

MCP access requires a Plus or Pro plan. Get your API key at Settings → API Keys before configuring.

Setup

Add WiseScribe to your MCP client config with the following:

MCP server config

{
  "wisescribe": {
    "url": "https://wisescribe.ai/api/mcp",
    "headers": {
      "Authorization": "Bearer wsc_YOUR_KEY"
    }
  }
}

Replace wsc_YOUR_KEY with your API key from Settings → API Keys.

Available tools

Once connected, your AI assistant has access to three WiseScribe tools:

Name
transcribe
Type
tool
Description
Transcribe a video or podcast URL. Supports YouTube, TikTok, Instagram, Facebook, X, Vimeo, Spotify, and Apple Podcasts. Returns the full transcript immediately for most platforms, or a jobId for platforms that process in the background.
Input: url (string, required) — the video or podcast URL to transcribe.
Name
poll_job
Type
tool
Description
Check the status of an async transcription job. Returns status (pending / processing / done / failed) and the full transcript when done.
Input: jobId (string, required) — the job ID returned by the transcribe tool.
Name
get_transcript
Type
tool
Description
Retrieve a saved transcript from your WiseScribe library by its ID.
Input: id (string, required) — the transcript ID from your WiseScribe account.

Claude Desktop

Open your Claude Desktop config file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

Add the WiseScribe server under mcpServers:

claude_desktop_config.json

{
  "mcpServers": {
    "wisescribe": {
      "url": "https://wisescribe.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer wsc_YOUR_KEY"
      }
    }
  }
}

Restart Claude Desktop. You'll see WiseScribe appear in the tools panel. You can now say things like:

"Transcribe this YouTube video: https://youtube.com/watch?v=..." "Summarize the podcast at this URL..."

Cursor

Open Cursor Settings → Features → MCP, then click Add new MCP server. Enter:

Name: wisescribe
Type: http
URL: https://wisescribe.ai/api/mcp
Headers: Authorization: Bearer wsc_YOUR_KEY

Or add it directly to .cursor/mcp.json in your project:

.cursor/mcp.json

{
  "mcpServers": {
    "wisescribe": {
      "url": "https://wisescribe.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer wsc_YOUR_KEY"
      }
    }
  }
}

Zed

In your Zed settings (~/.config/zed/settings.json), add:

Zed settings.json

{
  "context_servers": {
    "wisescribe": {
      "url": "https://wisescribe.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer wsc_YOUR_KEY"
      }
    }
  }
}

Usage tips

For YouTube and most podcasts, the transcribe tool returns the full transcript in one call
For TikTok, Instagram, and Facebook, use transcribe then call poll_job with the returned jobId
Transcripts count against your daily API quota — same as direct API calls
All transcripts created via MCP appear in your WiseScribe library at wisescribe.ai