Scrapegraph MCP

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/scrapegraphai-scrapegraph-mcp

About

Use cases: web-scraping, web-search

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list, install config and code. Checked Oct 25, 2026.

Tools

10 read · 6 write · 1 delete

Secrets

1 (API keys, tokens or passwords)

Install scripts

None found

Runs as

Python package

Tools (17)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

• crawl_get_statusread-only

  Retrieve the results of an asynchronous SmartCrawler operation. This tool fetches the results from a previously initiated crawling operation using the request_id. The crawl request processes asynchronously in the background. Keep polling this endpoint until the status field indicates 'completed'. While processing, you'll receive status updates. Read-only operation that safely retrieves results without side effects. Args: request_id: The unique request ID returned by crawl_start. Use this to retrieve the crawling results. Keep polling until status is 'completed'. Example: 'req_abc123xyz' …

• crawl_resume

  Resume a stopped crawl job (API v2 POST /crawl/:id/resume).

• crawl_start

  Start an asynchronous multi-page crawl (API v2 POST /crawl). Poll crawl_get_status with the returned id. Args: url: Starting URL (http/https). extraction_mode: Per-page output format — markdown (default), html, links, images, summary, branding, or screenshot. depth: Maximum crawl depth (0+). Maps to the v2 `maxDepth` field. Default 2. max_pages: Max pages (1-1000). Default 10. max_links_per_page: Max links to follow per page (>=1). Default 10. allow_external: Whether the crawler can cross domains. Default false. content_types: Allowed response content type…

• crawl_stop

  Stop a running crawl job (API v2 POST /crawl/:id/stop).

• creditsread-only

  Return remaining API credits (API v2 GET /credits).

• extractread-only

  Extract structured data from a webpage using AI (API v2 POST /extract). Args: user_prompt: Natural language instructions describing what data to extract. website_url: URL to extract data from (must include http:// or https://). output_schema: JSON schema (dict or JSON string) defining the expected output structure. If "required" field is missing, it will be automatically added as []. mode: Fetch/proxy mode — auto, fast, js. stealth: Use residential proxies to bypass bot detection (+5 credits). headers: Custom HTTP headers. cookies: Cookies to send with the …

• historyread-only

  List recent API requests (API v2 GET /history). v2 accepts page/limit/service. The legacy endpoint/offset params are kept as aliases and translated client-side (endpoint -> service, offset -> page when it divides limit). status filtering was removed in v2. Args: service: Filter by service name (e.g. scrape, extract, search). page: One-based page number. limit: Max rows per page (1–100). endpoint: Legacy alias for service. offset: Legacy offset; must be a non-negative multiple of limit.

• monitor_activityread-only

  Poll per-run tick history for a monitor (API v2 GET /monitor/:id/activity). Returns the ticks produced on each scheduled run (`id`, `createdAt`, `status`, `changed`, `elapsedMs`, `diffs`) plus `nextCursor` when more results are available. Mirrors `sgai.monitor.activity()` in scrapegraph-py v2. Args: monitor_id: ID of the monitor (cronId returned by monitor_create). limit: Page size, 1–100. Default 20 (server-side). cursor: Opaque pagination cursor returned as `nextCursor` by a prior call.

• monitor_create

  Create a scheduled monitor job (API v2 POST /monitor). The v2 API expects a `formats` array. This tool wraps `prompt` (+ optional `output_schema`) into a `{type: "json", ...}` format entry for you. Args: url: URL to monitor. prompt: Prompt describing what to extract on each run. interval: 5-field cron expression for scheduling. name: Optional monitor name. webhook_url: Optional webhook URL invoked when changes are detected. output_schema: JSON schema (dict or JSON string) for structured output. mode: Fetch/proxy mode — auto, fast, js. stealth: Use residential …

• monitor_deletecan change or delete data

  Delete a monitor (API v2 DELETE /monitor/:id).

• monitor_getread-only

  Get one monitor by id (API v2 GET /monitor/:id).

• monitor_listread-only

  List monitors (API v2 GET /monitor).

• monitor_pause

  Pause a monitor (API v2 POST /monitor/:id/pause).

• monitor_resume

  Resume a paused monitor (API v2 POST /monitor/:id/resume).

• schemaread-only

  Generate or augment a JSON Schema from a natural-language prompt (API v2 POST /schema). Args: prompt: Natural-language description of the schema to generate. existing_schema: Optional existing JSON Schema to extend (dict or JSON string). model: Optional LLM model override.

Compare and alternatives

Featured in

• Best MCP servers for data extraction

Signals

Stars

117

Forks

40

Open issues

2

Contributors

7

Weekly downloads

141

Last commit

Jul 17, 2026 (3 months ago)

Last release

Feb 7, 2026 (8 months ago)

Last package publish

Nov 19, 2025 (11 months ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name

io.github.ScrapeGraphAI/scrapegraph-mcp

Version

1.0.1

Auth

API key

Packages

pypi scrapegraph-mcp

License

MIT

Repository

github.com/ScrapeGraphAI/scrapegraph-mcp

First seen

Oct 24, 2026

Last checked

Oct 24, 2026

Source

Official MCP Registrygithub-graphql