Apple Podcasts MCP Server

Apple Podcasts Transcript MCP —
inside every chat.

Q: How do I connect it to Claude or Cursor?

Add a custom MCP connector pointing at https://mcp.transcriptmagic.com/mcp. Sign in with Google in the browser tab — the tools appear in every chat from then on.

Connect once, then ask Claude, ChatGPT, Cursor, or Claude Code to transcribe or summarize an Apple Podcasts episode. The AI calls transcribe_podcast, and for long episodes polls get_podcast_status until the transcript is ready.

MCP https://mcp.transcriptmagic.com/mcp

Connect in one click All MCP tools →

OAuth sign-in No API key 10 credits / audio-hour Tool: transcribe_podcast

Supported

What the tool handles.

Once connected, the AI just needs the URL — every Apple Podcasts format below resolves automatically. No URL cleanup, no parsing, no client-side handling.

Episode links

Apple Podcasts episode pages — a podcasts.apple.com URL with the ?i=<episodeId> parameter. Paste the specific episode, not the whole show.

AI speech-to-text

Podcasts rarely ship captions, so the transcript is generated with speech-to-text — you get text even when no captions exist.

Two-tool flow

Short episodes come back inline. For long ones, transcribe_podcast hands back a job_id and the AI polls get_podcast_status until it is done.

Full-length episodes

Multi-hour interviews and shows are handled — ideal for digesting a long episode without listening to the whole thing.

Prompts

Just ask.

After connecting, paste any prompt below into Claude, ChatGPT, or Cursor. The AI calls transcribe_podcast on its own — no syntax to memorize, no JSON to write.

Summarize

"Summarize this Apple Podcasts episode in five bullets: <episode URL>"

Show notes

"Turn this Apple Podcasts episode into chaptered show notes."

Repurpose

"Write a 600-word blog recap of this Apple Podcasts episode."

Q&A

"Read this episode and tell me what they say about the new product."

Clip ideas

"Find the five most quotable moments in this episode with rough timestamps."

Thread

"Convert the main argument from this episode into an X thread."

Setup

Connect in under a minute.

Paste https://mcp.transcriptmagic.com/mcp into your AI client's MCP / Connectors settings. A browser tab opens — sign in with Google, click Approve, done. Full per-client walkthroughs (Claude Desktop, Claude.ai, ChatGPT, Cursor, Claude Code) are on your dashboard.

One URL · OAuth · all clients

Open setup page →

Use cases

What people do with it.

The tool lives wherever you already write — chat windows, IDE side panels, agent pipelines. No glue code.

Show-note generation

Drop an episode into Claude and get chaptered notes, recaps, and quote callouts in one prompt.

Content repurposing

Episode → blog recap, interview → thread, show → newsletter — all in chat.

Research & digest

Get the gist of a long interview without listening to the whole thing.

Clip discovery

Ask Claude to surface the highest-signal moments worth cutting into short clips.

Quote mining

Pull direct quotes on a topic from a full-length episode with rough timestamps.

Cross-platform repurposing

Apple Podcasts episode → YouTube description or X thread in one prompt — same backend handles it.

FAQ

Questions people ask.

The short answers. For anything else, email [email protected] — same-day reply.

How do I connect it to Claude or Cursor?

Add a custom MCP connector pointing at https://mcp.transcriptmagic.com/mcp. Sign in with Google in the browser tab — the tools appear in every chat from then on.

Which plans can use podcast transcription?

It is available on the Plus and Pro plans. On the Free plan or a legacy one-time credit pack the tool returns an upgrade prompt. The social-video tools (YouTube, TikTok, etc.) stay available on every plan.

How much does it cost?

10 credits per audio-hour, reserved when the job starts and refunded automatically if it fails. A 30-minute episode is 10 credits; a 2-hour episode is 20.

Why does a long episode return a job_id instead of the transcript?

Long episodes take longer than a single tool call should block, so transcribe_podcast hands back a job_id after a short wait and the AI polls get_podcast_status until the transcript is ready. It is automatic from your side — just ask.

How accurate are the transcripts?

Transcripts are AI-generated speech-to-text. Clean spoken dialogue transcribes very well; heavy music, crosstalk, or poor audio can reduce accuracy on those portions.

Should I use this or the REST API?

Use the MCP server when you are inside Claude, ChatGPT, Cursor, or Claude Code. Use the Apple Podcasts REST API for batch processing and server-side automations. Same backend.

Connect Apple Podcasts transcripts — under a minute.

Included on the Plus and Pro plans. No API key. Same OAuth sign-in works across Claude Desktop, Claude.ai, ChatGPT, Cursor, and Claude Code.

Connect now See all tools

Apple Podcasts Transcript MCP — inside every chat.