Integration

Supadata integration

Supadata (formerly Gistly) turns videos and web pages into text an agent can use. Get transcripts of YouTube, TikTok, Instagram, X (Twitter) and Facebook videos or of a public media file, translate YouTube transcripts, read post, channel and playlist metadata, search YouTube, list a channel's or playlist's videos, run batch transcript and metadata jobs over many YouTube videos, extract structured data from a video with AI, and scrape, map or crawl websites into Markdown. Long jobs (large transcripts, extracts, batches, crawls) return a job id that the matching get tool reads back. The two YouTube-only video and transcript routes the vendor marks deprecated are not exposed: use get_media_metadata and get_transcript for YouTube videos instead. Authentication is an API key sent in the x-api-key header.

Connect Supadata How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Supadata.Built from the 19 Supadata tools below, on the connection you add once.

What you can do with Supadata

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Crawl, 2 operations

create_crawlCreate crawlPOSTStart a crawl job that extracts the Markdown content of a website's pages; set limit (when left out, up to 100 pages at 1 credit each). Returns jobId: read the pages with get_crawl.
get_crawlGet crawlGETRead a crawl job started by create_crawl. status is scraping, completed, failed or cancelled; pages holds the crawled pages once completed. When next is present, call again with skip to get the next page.

Extract, 2 operations

create_extractCreate extractPOSTStart an AI job that extracts structured data from a video URL, guided by a prompt, a JSON Schema, or both (at least one). Returns jobId: read the result with get_extract.
get_extractGet extractGETRead an extract job started by create_extract. status is queued, active, completed or failed; data holds the extracted result when completed, error when failed, and schema the AI-generated schema when none was sent.

Transcript, 2 operations

get_transcriptGet transcriptGETAnswers the transcript in content or, for a large video, only jobId: then call get_transcript_job with that jobId until status is completed or failed. Works for YouTube, TikTok, Instagram, X (Twitter) and Facebook videos and public media file URLs.
translate_youtube_transcriptTranslate youtube transcriptGETTranslate the transcript of a YouTube video into the language given in lang. Pass the video as url or as videoId (one of the two).

Account, 1 operation

get_accountGet accountGETGet the Supadata account behind the API key: organization id, plan name, and the credits available and used in the current billing period. Not rate limited.

Batch, 1 operation

get_youtube_batchGet youtube batchGETRead a YouTube batch job started by create_youtube_transcript_batch or create_youtube_video_batch. status is queued, active, completed or failed; results holds one entry per video. Page with limit and offset.

Channel, 1 operation

get_youtube_channelGet youtube channelGETGet a YouTube channel's metadata (name, handle, description, subscriber, video and view counts, thumbnail, banner) from its URL, handle or ID.

Channel video, 1 operation

list_youtube_channel_videosList youtube channel videosGETList the video IDs of a YouTube channel, split into regular videos, Shorts and live streams. Filter with type. Pass an ID to get_media_metadata or get_transcript for details.

Media metadata, 1 operation

get_media_metadataGet media metadataGETGet the metadata of a YouTube, TikTok, Instagram, X (Twitter) or Facebook post or video from its URL: title, description, author, engagement stats, media details, tags and creation date, in one shape for every platform.

Playlist, 1 operation

get_youtube_playlistGet youtube playlistGETGet a YouTube playlist's metadata (title, description, video and view counts, last update, owning channel) from its URL or ID.

Playlist video, 1 operation

list_youtube_playlist_videosList youtube playlist videosGETList the video IDs of a YouTube playlist, split into regular videos, Shorts and live streams. Pass an ID to get_media_metadata or get_transcript for details.

Transcript batch, 1 operation

create_youtube_transcript_batchCreate youtube transcript batchPOSTStart a batch job that gets the transcripts of many YouTube videos. Give exactly one of videoIds, playlistId or channelId. Returns jobId: read the results with get_youtube_batch.

Transcript job, 1 operation

get_transcript_jobGet transcript jobGETRead a transcript job started by get_transcript, by its jobId. status is queued, active, completed or failed; content, lang and availableLangs are filled when completed, error when failed. Results stay available 1 hour.

Video batch, 1 operation

create_youtube_video_batchCreate youtube video batchPOSTStart a batch job that fetches the metadata of many YouTube videos. Give exactly one of videoIds, playlistId or channelId. Returns jobId: read the results with get_youtube_batch.

Web page, 1 operation

scrape_web_pageScrape web pageGETScrape one web page into Markdown: returns its content, name, description, Open Graph URL, character count and the URLs it links to. Use map_website for a whole site's URLs, create_crawl for all its pages.

Website, 1 operation

map_websiteMap websiteGETList the URLs found across a whole website, for example to build a sitemap or choose pages to scrape. Returns urls only; use scrape_web_page or create_crawl for content.

Youtube result, 1 operation

search_youtubeSearch youtubeGETSearch YouTube for videos, channels, playlists or movies, with upload date, duration, feature and sort filters. Without limit one page is returned with nextPageToken; with limit the API pages for you (more credits).

Use Supadata from Claude Code, Cursor or Codex

Connect Supadata in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Supadata operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Supadata

Create your workspace
Sign up free. Nothing to install.
Connect Supadata once
Add your Supadata connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.