Peep Is The API To Scrape, Search, And Extract From Any Website. Built For Agents, RAG, And Crawls.
{
"url": "https://example.com",
"title": "Example Domain",
"markdown": "# Example Domain\n\nThis Domain...",
"links": [
"https://www.iana.org/domains"
],
"status": 200
}The Infrastructure Layer That Helps AI Find, Read, And Act On The Live Web.
Search The Web And Get Full Content From Results.
Get LLM-Ready Data From Websites. Markdown, JSON, Screenshot, And More.
Pull Typed JSON From Any URL With A Schema Or A Natural-Language Prompt.
1# pip install peep2from peep import Peep34peep = Peep(api_key="peep_live_...")56# Scrape a website:7peep.scrape("peep.dev")1# Peep23Peep Helps AI Systems Search, Scrape,4And Extract From Any Website.56## Features78- Search: Find Information Across The Web9- Scrape: Clean Data From Any Page10- Crawl: Walk Whole Sites With Filters11- Extract: Typed JSON With A SchemaConnect Peep To Any AI Agent Or MCP Client In Minutes.
Install The JavaScript Or Python SDK And Start Calling Scrape, Crawl, Or Extract From Your Agent Loop.
1npm i @peep/sdk2# or3pip install peepAdd Peep As A Model Context Protocol Server To Claude Desktop, Cursor, Or Any MCP Client.
1{2 "mcpServers": {3 "peep-mcp": {4 "command": "npx",5 "args": ["-y", "@peep/mcp"],6 "env": {7 "PEEP_API_KEY": "peep_live_..."8 }9 }10 }11}1curl -s https://peep.dev/skill/ONBOARD.mdWeb Data Infrastructure Built From The Ground Up.
Everything You Need To Know About Peep.
Peep Is An API That Turns Any Web Page Into Clean Markdown, A Screenshot, Or Structured JSON — With JS Rendering, Proxies, Retries, And Caching Handled For You.
RAG Pipelines, Lead Enrichment Tools, Price And Inventory Trackers, Deep-Research Agents, Content Monitors, And Any App That Needs Fresh Structured Data From The Web.
LLMs Can Read Text, But They Can't Fetch Pages, Bypass Bot Protection, Or Turn Messy HTML Into Clean JSON. Peep Handles All Of That Behind One Call So The Agent Can Focus On Reasoning.
Scrape Turns One URL Into Clean Data. Search Queries The Web And Optionally Pulls Full Content. Crawl Walks A Whole Site Under Filters. Extract Returns Typed JSON From One Or Many URLs With A Schema Or Prompt.
Yes. We Ship A Typed JS SDK, A Python SDK, A CLI, And An MCP Server So Claude, Cursor, Or Your Custom Agent Can Call Peep Tools Directly.
Yes, By Default On Every Plan. Paid Tiers Can Opt Out Per Request With A Flag; Overrides Are Audit-Logged.
Yes — 500 Credits On Sign-Up, No Card Required. One Credit Per Page Covers Most Scrapes.