Account

Integrations

MCP server

Give an assistant the ability to read Instagram Reels. One URL, browser sign-in, and the tools it exposes.

Assistants cannot watch video. Ask one about an Instagram link and it will either refuse or answer from the caption, which is usually wrong.

The MCP server fixes that. Hand it a Reel link, get the spoken words back. From there the model can summarise it, pull the claims out of it, or write from it.

What MCP is

Model Context Protocol is a standard way to hand tools to an assistant. Claude, Cursor, ChatGPT and a growing number of other clients speak it. You add a server once and every conversation can use it.

Connect it

There is nothing to install. The server is hosted, and any client that speaks remote MCP over Streamable HTTP with OAuth can use this address:

https://instagramtotranscript.com/api/mcp

The first time a client connects, your browser opens and asks you to sign in and approve it. After that the connection stays until you disconnect it from your account page.

Claude Code

claude mcp add --transport http itt https://instagramtotranscript.com/api/mcp

Claude Desktop and Claude on the web

Add a custom connector, paste the server URL above, and pick OAuth when asked how it should authenticate.

Cursor

Settings, then Tools and MCP, then add a custom server with that URL.

ChatGPT

Turn on developer mode, then create a custom app with the server URL and OAuth authentication.

Headless clients

A script or a container has no browser to sign in with. Send an API key from your account page as a bearer token instead, and skip OAuth entirely.

{
  "mcpServers": {
    "itt": {
      "type": "http",
      "url": "https://instagramtotranscript.com/api/mcp",
      "headers": { "Authorization": "Bearer itt_your_key_here" }
    }
  }
}

The tools

transcribe_reel

Takes a public Instagram link. Returns the transcript, the detected language, the duration and the account handle.

Argument Type Required
url string yes
language string no
withTimestamps boolean no

Set withTimestamps when the model needs to point at a moment in the video. Leave it off for summarising, where the timings only add noise.

transcribe_many

Takes up to twenty links and returns one result per link. Failures come back marked rather than throwing, so one private post does not lose the whole batch.

Use this for competitor sweeps. It is far cheaper in tokens than twenty separate calls.

get_credit_balance

How many credits are left. Worth asking before a large batch, and the model will tell you when a run is going to stop short.

What it costs

One credit is one minute of audio, rounded up. The same balance covers the API and the MCP server, so an assistant and your own code draw on one pot. New accounts start with 10 credits. The pricing page has the packs.

A batch that runs out of credits marks the remaining links rather than failing silently, so the model can tell you exactly where it stopped.

Things worth asking it

  • "Transcribe this Reel and pull out every claim as a bullet list."
  • "Read these five Reels and tell me what they all agree on."
  • "Transcribe this and draft a newsletter opening in my voice."
  • "What question does this Reel answer? Write three headlines for it."

Limits

The same rules as the API. Public posts only, video up to ten minutes, and a rate limit per key.

Private posts come back as an error the model can read, so it will tell you the post was unreadable rather than inventing a transcript.

Without an account

The free tool does the same job one link at a time in a browser. Paste a transcript into your assistant by hand to see whether the workflow is worth automating before you connect anything.