Integration

ScrapeGraphAI integration

ScrapeGraphAI (v2 API) turns web pages into data for AI workflows. Scrape a URL into markdown, HTML, links, images, a summary, a screenshot or LLM-extracted JSON; extract structured JSON from a URL, raw HTML or markdown with a natural-language prompt; run a web search that fetches the top results; start, poll, stop, resume and read multi-page crawls; schedule cron monitors with change detection; and read request history and the credit balance. Every call except history, credits and status reads spends credits. The internal worker routes, the crypto-wallet credit purchase and agent-payment routes, the agent enrollment (AEP/ODP) routes and the health check are not exposed. Authentication is an API key sent in the SGAI-APIKEY header.

Connect ScrapeGraphAI How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in ScrapeGraphAI.Built from the 21 ScrapeGraphAI tools below, on the connection you add once.

What you can do with ScrapeGraphAI

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.

Monitor, 7 operations

create_monitorCreate monitorPOSTCreates a monitor that fetches a URL on a cron schedule, captures it in the given formats and records change diffs between ticks, optionally posting each tick to a webhook. Every tick spends credits.
delete_monitorDelete monitorDELETEPermanently removes a monitor and its tick history.
get_monitorGet monitorGETReturns one monitor with its resolved config, schedule and status.
list_monitorsList monitorsGETLists the account's monitors with their cronId, status (active or paused), schedule and config.
pause_monitorPause monitorPOSTPauses a monitor: future ticks stop, the monitor record is kept. resume_monitor puts it back on schedule.
resume_monitorResume monitorPOSTResumes a paused monitor, putting it back on schedule immediately. Ticks spend credits again.
update_monitorUpdate monitorPATCHUpdates a monitor's name, interval, formats, webhookUrl or fetchConfig. Send only the fields to change; leave a field out to keep its current setting.

Crawl, 5 operations

delete_crawlDelete crawlDELETEPermanently removes a crawl record and its stored pages. Cannot be undone.
get_crawlGet crawlGETReturns the progress of a crawl (status, total, finished) and lightweight per-page metadata. Poll every few seconds until status is no longer running; page content comes from list_crawl_pages.
resume_crawlResume crawlPOSTResumes a stopped crawl from its last frontier. Further pages spend credits.
start_crawlStart crawlPOSTStarts an async multi-page crawl and returns its job id at once; poll get_crawl with that id until status is no longer running, then read content with list_crawl_pages. Costs 2 credits plus each page.
stop_crawlStop crawlPOSTStops an in-flight crawl. Already-fetched pages stay available and no further URLs are expanded; resume_crawl continues it.

History entry, 2 operations

get_history_entryGet history entryGETReturns one past request with its full result payload. Pass the id from scrape_url, extract_data or search_web, a crawl page's scrapeRefId, or a monitor tick id.
list_historyList historyGETLists past requests of the account (scrape, extract, search, monitor ticks, crawl jobs), newest first, with their params and result. Filter by service or sessionId; paginate with page and limit.

Api key, 1 operation

validate_api_keyValidate api keyGETChecks the API key and returns the email of the user who owns it.

Crawl page, 1 operation

list_crawl_pagesList crawl pagesGETReturns a slice of a crawl's pages with each page's scrape result resolved under scrape. Pass pagination.nextCursor as cursor for the next slice; null means no more pages.

Credits, 1 operation

get_creditsGet creditsGETReturns the remaining credit balance, credits used this billing cycle, the plan name and crawl and monitor job slots. Costs no credits.

Data, 1 operation

extract_dataExtract dataPOSTExtracts structured JSON with a natural-language prompt (and optional JSON schema) from exactly one source: url, html or markdown. Costs 5 credits, plus 5 with stealth. Result is under json.

Monitor tick, 1 operation

list_monitor_activityList monitor activityGETReturns a monitor's recent ticks with status, elapsed time, a changed flag and per-format diffs. Pass nextCursor back as cursor for older ticks; a tick id works with get_history_entry.

Page, 1 operation

scrape_urlScrape urlPOSTFetches one public URL and returns its content in the formats you pick (markdown, html, links, images, summary, json, branding, screenshot) in one call. Spends credits per format; stealth adds 5.

Web, 1 operation

search_webSearch webPOSTRuns a web search and fetches the top results with their content inline; add a prompt (and schema) to also get one structured json across them. Costs 2 credits per result, 5 with a prompt.

Use ScrapeGraphAI from Claude Code, Cursor or Codex

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

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