Integration

DumplingAI integration

DumplingAI is a web data API for AI agents. An agent can search Google (web, news, places, maps, autocomplete), read Google reviews of a business, scrape, crawl, screenshot or AI-extract web pages, read YouTube channels, videos, comments, searches and transcripts, read TikTok profiles, videos, comments and transcripts, read a LinkedIn company page, extract data from documents, images, audio and video with a prompt, convert, merge or tag PDFs, trim a video and generate AI images. Every call spends credits from the plan (each tool states its cost; failed calls are free; 403 means not enough credits); plans include 30 to 120 requests per minute. Not exposed: LinkedIn person profiles, TikTok user search and TikTok follower and following lists (research about individuals), the agent completion route (a dashboard agent's tools reach the excluded routes), the code-execution sandboxes, the undocumented extract-pdf route, the very large Google locations list, and the unified v2 run API. Commenter and reviewer identities are not returned. Authentication is a DumplingAI API key sent as a Bearer token.

Connect DumplingAI How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in DumplingAI.Built from the 34 DumplingAI tools below, on the connection you add once.

What you can do with DumplingAI

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.

Pdf, 2 operations

convert_to_pdfConvert to pdfPOSTConverts a document, image or web file to PDF and returns a temporary link to it (kept about 24 hours). Costs 50 credits per 100 MB of output PDF.
merge_pdfsMerge pdfsPOSTMerges two or more PDFs into one and returns a temporary link to it (kept about 24 hours). Costs 50 credits per 100 MB of output PDF.

Pdf metadata, 2 operations

read_pdf_metadataRead pdf metadataPOSTReads the standard metadata fields (title, author, dates...) of PDF files. Costs 10 credits per request.
write_pdf_metadataWrite pdf metadataPOSTWrites metadata fields into PDFs and returns a temporary link to the new file (a ZIP for several PDFs, kept about 24 hours). Your originals are not changed. Costs 50 credits per 100 MB of output.

Tiktok video, 2 operations

get_tiktok_videoGet tiktok videoPOSTGets a TikTok video's details: caption, statistics, author, music and hashtags. Costs 10 credits per request.
list_tiktok_profile_videosList tiktok profile videosPOSTLists the videos posted by a TikTok account with statistics. Pages with maxCursor while has_more is 1. Costs 10 credits per request.

Youtube video, 2 operations

get_youtube_videoGet youtube videoPOSTGets a YouTube video or short: title, description, counts, publish date, duration, channel, suggested videos and, on request, the transcript. Costs 10 credits per request.
list_youtube_channel_videosList youtube channel videosPOSTLists a YouTube channel's long-form videos with views and publish dates. Pages with continuationToken. Costs 10 credits per request.

Audio insight, 1 operation

extract_audioExtract audioPOSTAnalyses an audio file with an AI model following your prompt (summary, transcript, structured fields). Costs 100 credits plus 20 per minute of audio.

Autocomplete suggestion, 1 operation

get_autocompleteGet autocompletePOSTReturns Google autocomplete suggestions for a query. Costs 30 credits per request.

Document data, 1 operation

extract_documentExtract documentPOSTExtracts information from documents (PDF, Word, spreadsheets, slides and more) with a vision AI model following your prompt. Costs 100 credits plus 10 per page.

Document text, 1 operation

doc_to_textDoc to textPOSTConverts a PDF or DOCX document to plain text. Costs 20 credits per request.

Google review, 1 operation

get_google_reviewsGet google reviewsPOSTGets Google reviews of a business (found by keyword, Google CID or Place ID): rating, review text, owner answers. Costs 50 credits per 10 reviews, rounded up.

Image, 1 operation

generate_ai_imageGenerate ai imagePOSTGenerates images from a text prompt. Billed per image by model: FLUX.1-schnell 30 credits, FLUX.1-dev 150, FLUX.1-kontext-pro and recraft-v3 200, FLUX.1-pro 300, FLUX.1.1-pro 500.

Image data, 1 operation

extract_imageExtract imagePOSTReads images (OCR and visual content) with a vision AI model following your prompt. Costs 100 credits plus 10 per image.

Linkedin company, 1 operation

get_linkedin_companyGet linkedin companyPOSTGets a LinkedIn company page: description, industry, size, headquarters, website, followers and recent posts. Employee lists are not returned. Costs 10 credits per request.

Map place, 1 operation

search_mapsSearch mapsPOSTSearches Google Maps and returns places with address, coordinates, type, website, opening hours and IDs. Costs 100 credits per page.

News article, 1 operation

search_newsSearch newsPOSTSearches Google News and returns articles with title, source, date, snippet and link. Costs 30 credits per page.

Place, 1 operation

search_placesSearch placesPOSTSearches Google Places for businesses and points of interest: address, coordinates, rating, category, phone and website. Costs 30 credits per request.

Screenshot, 1 operation

screenshot_webpageScreenshot webpagePOSTTakes a screenshot of a web page and returns a temporary image link. Costs 20 credits, plus 10 when autoScroll is true and 10 when wait is set.

Tiktok comment, 1 operation

list_tiktok_video_commentsList tiktok video commentsPOSTLists comments and their replies on a TikTok video (text, likes, time; commenter identities are not returned). Pages with cursor while hasMore. Costs 10 credits per request.

Tiktok profile, 1 operation

get_tiktok_profileGet tiktok profilePOSTGets a TikTok account's public profile (name, bio, verified flag) and statistics (followers, likes, videos). Costs 10 credits per request.

Tiktok transcript, 1 operation

get_tiktok_transcriptGet tiktok transcriptPOSTGets the transcript of a TikTok video in WebVTT format. Costs 10 credits per request.

Video clip, 1 operation

trim_videoTrim videoPOSTCuts an MP4 video between two timestamps and returns a temporary link to the clip (kept about 24 hours, up to 200 MB). Costs 200 credits per minute of output, rounded up.

Video insight, 1 operation

extract_videoExtract videoPOSTAnalyses a video with an AI model following your prompt. Billed per SECOND: 100 credits plus 10 per second of video (a 1-minute clip costs 700).

Web result, 1 operation

search_webSearch webPOSTRuns a Google web search and returns organic results, featured snippet, related searches and People also ask. Costs 30 credits per page; scraping results is billed extra (10 each).

Webpage, 1 operation

scrape_webpageScrape webpagePOSTScrapes one web page and returns its content as markdown, HTML or a screenshot, with title and metadata. Costs 10 credits per request.

Webpage data, 1 operation

extract_webpage_dataExtract webpage dataPOSTExtracts data matching your JSON schema from a web page, using its screenshot and HTML with an AI model. Returns one object. Costs 250 credits per request.

Website, 1 operation

crawl_websiteCrawl websitePOSTCrawls a website from a start URL and returns the content of each page. Costs 10 credits per crawled page; limit caps the pages (5 when left out).

Youtube channel, 1 operation

get_youtube_channelGet youtube channelPOSTGets a YouTube channel's details: name, description, subscriber, video and view counts, join date, country and social links. Pass channelId, handle or url. Costs 10 credits.

Youtube comment, 1 operation

list_youtube_video_commentsList youtube video commentsPOSTLists comments on a YouTube video (text, time, likes, replies, nesting; commenter identities are not returned). Pages with continuationToken. Costs 10 credits per request.

Youtube result, 1 operation

search_youtubeSearch youtubePOSTSearches YouTube for videos, shorts, channels, playlists or live streams. Results come in one list per type; pages with continuationToken. Costs 10 credits per request.

Youtube short, 1 operation

list_youtube_channel_shortsList youtube channel shortsPOSTLists a YouTube channel's Shorts with title, thumbnail and views. Pages with continuationToken. Costs 10 credits per request.

Youtube transcript, 1 operation

get_youtube_transcriptGet youtube transcriptPOSTGets the transcript of a YouTube video, optionally with timestamps. Falls back to the first available language when the preferred one is missing. Costs 20 credits per request.

Use DumplingAI from Claude Code, Cursor or Codex

Connect DumplingAI in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its DumplingAI 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 DumplingAI

Create your workspace
Sign up free. Nothing to install.
Connect DumplingAI once
Add your DumplingAI 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.