Scrapi
Turn arbitrary URLs into clean Markdown in your coding agent while bypassing common anti-bot blocks via Scrapi's hosted MCP.
Overview
Scrapi is a MCP server for the Build phase that converts URLs to clean, LLM-ready Markdown for AI agents with anti-bot bypass via Scrapi's API.
What is this MCP server?
- Converts URLs to LLM-ready Markdown for agent context windows
- Anti-bot bypass handled by Scrapi hosted service
- Remote streamable HTTP at scrapi.ai/mcp with Bearer API key
- npm package @scrapi.ai/mcp-server v2.0.6 with stdio transport option
- Package version 2.0.6
- Remote MCP URL https://scrapi.ai/mcp
- npm identifier @scrapi.ai/mcp-server
Community signal: 5 GitHub stars.
What problem does it solve?
Your agent needs readable page content but raw fetch and bot challenges waste time and break pipelines.
Who is it for?
Indie builders adding web research, RAG ingestion, or competitive monitoring to agents without self-hosting a scraper farm.
Skip if: Projects that require on-prem only scraping, heavy JavaScript SPAs without API support, or large-scale crawl compliance programs without your own legal review.
What do I get? / Deliverables
You register Scrapi MCP and pass URLs so tools return Markdown your agent can cite, summarize, or index immediately.
- Clean Markdown snapshots of target URLs inside agent sessions
- Repeatable MCP tool calls via remote or stdio transport
Recommended MCP Servers
Journey fit
Fetching and normalizing web content is a build-time integration task when you wire research, RAG, or monitoring into the product. Integrations is the shelf because Scrapi is an external scraping API exposed as MCP tools, not a ship-phase test runner or launch SEO workflow.
How it compares
Hosted scrape-to-Markdown MCP integration, not a browser automation skill or local Playwright script pack.
Common Questions / FAQ
Who is Scrapi for?
Solo developers and agent authors who want URL-to-Markdown fetching with anti-bot handled by a managed API.
When should I use Scrapi?
Use it while building integrations that need fresh web pages inside Claude Code, Cursor, or Codex workflows.
How do I add Scrapi to my agent?
Get a free API key at https://scrapi.ai/dashboard, then add the remote MCP https://scrapi.ai/mcp with Authorization Bearer SCRAPI_API_KEY or install @scrapi.ai/mcp-server and set SCRAPI_API_KEY for stdio.