TrueMemory Web / Built for agent loops

Give your agents a live web memory.

Search, extract, map, and ground an answer in sources your team can inspect.

Live web result

Ready for a URL or query

Your live web result will appear here

Enter a URL or search query above. TrueMemory Web will return a readable answer, source links, and the raw response.

Features

Search, extract, map, and crawl

Run each tool on its own, or ground an answer in the sources they return.

Search

Search the web with AI-powered understanding. Find relevant content across millions of pages.

Crawl

Intelligent crawling that respects robots.txt while discovering all important pages on a site.

Scrape

Extract clean markdown, structured data, or screenshots from any webpage at scale.

Map

Discover the full structure of a website. Map all pages and their relationships.

API & SDK

REST API, TypeScript SDK, Python SDK, and MCP integration for any workflow.

Browser Automation

Handle JavaScript-heavy sites, SPAs, and dynamic content with headless browser support.

Use Cases

Current web context for your agents

Feed current, inspectable web sources into RAG pipelines and agent workflows.

RAG Applications

Feed clean, structured web data into your retrieval-augmented generation pipeline.

AI Agents

Give your agents the ability to search, browse, and extract information from the web.

Topic Research

Monitor competitors, track changes, and gather intelligence at scale.

Content Extraction

Turn messy web pages into clean, structured data for analysis and storage.

Developer-first

One API for every web task

TrueMemory Web exposes search, scrape, map, crawl, and agent extraction through REST, TypeScript, Python, and MCP.

TypeScriptPython
import { AmanCrawl} from '@AmanCrawl/sdk';

const client = new AmanCrawl({
  apiKey: process.env.AmanCrawl_API_KEY,
});

// Scrape a webpage
const result = await client.scrape({
  url: 'https://example.com',
  formats: ['markdown', 'json'],
});

console.log(result.markdown);
// # Page Title
// Content extracted...

console.log(result.json);
// { title: "...", content: "..." }

Give your agent current web context

Start with 1,000 scraped pages per month. No credit card required.