What it does
Audivo is an MCP server that returns podcast transcripts. A client sends an episode link, an episode ID, or an uploaded file, and receives the transcript in pages of about 40,000 characters. Other tools search shows, list episodes and charts, price a batch of episodes before anything runs, track transcription groups, and list recent transcripts. Clients that render MCP Apps display transcripts in a reader with timestamps and search.
It suits anyone who wants to transcribe an episode and summarise it from inside Claude, ChatGPT, Codex or Cursor. For many episodes at once, such as a chart or a back catalogue, the quote tool gives a total before the confirm tool starts the jobs. The local server also accepts YouTube links and audio files from the user's machine.
The hosted endpoint runs on Audivo's servers and is reached by URL, with OAuth sign-in or a bearer key. The local server runs through npx on the user's machine and reads an API key from the AUDIVO_API_KEY environment variable set in the client. Transcription spends credits, roughly one per audio minute, and the max_credits parameter refuses a call that could cost more. Repeating the same call returns the existing job rather than charging again.
Limitations and requirements
An Audivo API key from the dashboard is required; keys are shown only once. Transcription spends credits, about one per audio minute, and a batch job holds up to the quote's ceiling. YouTube links, local file paths and file upload work only on the local server. The local version runs through npx, which checks npm at each client start when @latest is used. Keep files that contain a real key out of Git and shared chats.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
Audivo
Podcast transcripts: send an episode link and get the transcript back, a page at a time.
v0.4.2As shown on registry.modelcontextprotocol.io when we last read it