Dictationer for Cursor

Give Cursor's agent access to video transcripts

Add one MCP server and let Cursor turn product demos, talks, and launch videos into captions, documentation, and changelog entries.

Get my connection linkSee examples

Works on every plan, including Free

ChatGPTClaudeCursorVS CodeClaude Code

Add captions to your project

You

Create English captions for this product demo and save them to public/captions/demo.vtt:

https://www.youtube.com/watch?v=…

Used Dictationer

transcribe_url({ url, output_format: "vtt" })

Assistant

I saved the captions to public/captions/demo.vtt and added a <track> element to the demo video player.

public/captions/demo.vtt

WEBVTT

00:00:00.000 --> 00:00:03.200
Let's export a report in under ten seconds.

00:00:03.200 --> 00:00:06.900
Pick a date range, then choose the format you need.
User 1User 2User 3User 4User 5

210k+ users trust Dictationer

Add Dictationer to Cursor

  1. 1

    In Dictationer, open Dashboard → Integrations and copy your AI assistant connection link.

  2. 2

    In Cursor, open Settings → MCP and add a new MCP server, or edit ~/.cursor/mcp.json.

  3. 3

    Paste the configuration below, replacing the placeholder with your link, and save.

~/.cursor/mcp.json

{
  "mcpServers": {
    "dictationer": {
      "url": "YOUR_DICTATIONER_LINK"
    }
  }
}

Examples

What you can ask

Example conversations. Dictationer fetches the transcript; your AI assistant does the rest.

Draft release notes from a launch video

You

Use this launch video to draft entries for CHANGELOG.md under a new version:

https://x.com/…/status/…

Used Dictationer

transcribe_url({ url })

Assistant

Added to CHANGELOG.md: • New: scheduled exports to email and Slack. • Improved: reports load about twice as fast. • Fixed: date filters now respect your time zone.

Create YouTube chapters

You

Create chapter timestamps for this video's description:

https://www.youtube.com/watch?v=…

Used Dictationer

transcribe_url({ url, output_format: "timestamped" })

Assistant

Paste these into your video description:

Chapters

0:00 Intro
1:12 Why written decisions matter
4:30 The weekly review
7:05 Tools we use
9:40 Q&A

Works with public video links from

YouTubeYouTube ShortsTikTokInstagram ReelsX (Twitter)FacebookThreadsLinkedIn

Supported apps

Use it wherever you work with AI

Dictationer works with any AI app that supports remote MCP servers.

ChatGPT

Summaries, notes, and translations of video links right in ChatGPT.

Set up

Claude

Add Dictationer as a custom connector in Claude on the web or desktop.

Set up

VS Code

Use video transcripts in GitHub Copilot agent mode.

Set up

Claude Code

Add Dictationer to Claude Code with one terminal command.

Set up
Dictationer MCP

Your link stays private

Your connection link is personal to your Dictationer account.

Replace or disconnect it any time from Dashboard → Integrations.

Dictationer only fetches the public video links you share in the chat.

FAQ

Frequently Asked Questions

Add a server named dictationer with your connection link as its url in ~/.cursor/mcp.json, or add it from Cursor Settings → MCP.

Yes. Once the server is added, the agent can call Dictationer whenever you share a video link.

No. You copy one private link from Dashboard → Integrations in Dictationer and paste it into your AI app's connector settings.

Yes. On the free plan, each transcription uses your free Dictationer credits, the same as transcribing a link on the website. Paid plans include transcription from AI assistants without using credits.

Public videos from YouTube, YouTube Shorts, TikTok, Instagram Reels, X (Twitter), Facebook, Threads, and LinkedIn.

Try it in your next chat

Create your connection link and ask your first question in under a minute.

Get my connection linkSee plans