Atlas
OfficialWhat Atlas Does (Before & After)
Input (Messy Web HTML) | Atlas Output (LLM-Ready Markdown) |
• 1.8 MB raw DOM soup• 240,000+ tokens• Cookie modals, tracking pixels, ads, navigation clutter• Deeply nested | • 12 KB high-density GitHub Flavored Markdown• 2,800 tokens (98.8% token reduction)• Preserved tables, clean code fences, and YAML frontmatter• 100% deterministic, zero LLM hallucinations |
⚡ Why Atlas?
Traditional web scraping returns megabytes of raw, unstructured HTML bloated with navigation bars, cookie banners, tracking scripts, and presentational CSS. Feeding raw HTML into Large Language Models wastes 70% to 90% of your context window, escalates inference costs, and introduces reasoning hallucinations.
Atlas compiles web content into clean, token-efficient markdown at the edge—before it reaches your models:
🧠 Deterministic by Design: Rule-based AST heuristic passes produce identical, reliable output for identical inputs. Zero prompt drift, zero hallucinated content, and zero LLM cost during compilation.
📉 70–90% Token Reduction: Eliminates boilerplate chrome, banners, and layout bloat while faithfully preserving headings, tables, code fences, equations, and semantic links.
🤖 Built for AI Agents: Native Model Context Protocol (MCP) server for instant integration into Claude Desktop, Cursor, VS Code, and Windsurf, plus native SDKs (TypeScript, Python, Go) for custom agent loops.
⚡ Sub-1.5s Edge Execution: Powered by Cloudflare Workers across 300+ edge locations with built-in DNS-aware SSRF protection.
🔁 Safe Retries & Idempotency: Built-in request deduplication ensures retrying failed or timed-out requests never burns duplicate compute credits or double-executes jobs.
🌐 Engineered for Real-World Content: High-fidelity extraction tailored for documentation engines (Mintlify, Docusaurus, GitBook), technical blogs, research publications, and modern SSR/SSG sites.
The result: High-signal, token-efficient input for AI agents and RAG pipelines—without spending LLM tokens just to clean up HTML.
Related MCP server: PageMap
📦 What’s in This Repository
This monorepo houses the public ecosystem and developer tools for Atlas:
Component | Path | Language / Standard | Installation / Package |
Model Context Protocol | MCP Standard (2026-07-28) |
| |
TypeScript / JS SDK | TypeScript / Node / Deno / Bun |
| |
Python SDK | Python 3.10+ (httpx, typing) |
| |
Go SDK | Go 1.22+ |
| |
Atlas Agent Skills | Agent Skill Specification | Canonical recipes for AI assistants | |
Starter Examples | Node, Python, Go, cURL | Runnable recipes & templates |
🚀 Quickstarts
🔑 Authentication & API Key Setup
Get your API Key: Sign in to the Atlas Dashboard to generate your free API key (
atlas_...).Set your Environment Variable (Recommended): All Atlas SDKs, MCP servers, and CLI tools automatically detect
ATLAS_API_KEY:# macOS / Linux (add to ~/.bashrc or ~/.zshrc) export ATLAS_API_KEY="atlas_your_api_key_here" # Windows PowerShell $env:ATLAS_API_KEY = "atlas_your_api_key_here"
Prefer not to set environment variables? You can pass your API key directly into any SDK client constructor as shown below.
Direct HTTP / cURL
# Instant Real-Time Scrape (Sub-1.5s, clean markdown output)
curl -X POST https://api.atlas-compiler.com/v1/scrape \
-H "Authorization: Bearer ${ATLAS_API_KEY:-atlas_your_api_key_here}" \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com"}'
# Deep Semantic Compilation (includes full AST, link graph, and idempotency key)
curl -X POST https://api.atlas-compiler.com/v1/compile \
-H "Authorization: Bearer ${ATLAS_API_KEY:-atlas_your_api_key_here}" \
-H "Idempotency-Key: $(uuidgen)" \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com"}'AI Agents & MCP Server
Connect Claude Desktop, Cursor, VS Code, or Windsurf directly to Atlas with the 1-click interactive wizard:
npx -y @atlascompiler/mcp initThe wizard prompts for your Atlas API key, validates connectivity, and automatically configures your installed editors.
To run the Stdio server directly:
npx -y @atlascompiler/mcpFor Remote / Cloud AI Agents (Streamable HTTP):
Endpoint:
https://api.atlas-compiler.com/mcpHeaders:
Authorization: Bearer atlas_your_api_key_here
TypeScript / JavaScript
Install the official type-safe client library:
npm install @atlascompiler/sdk
# or: pnpm add @atlascompiler/sdk / yarn add @atlascompiler/sdkimport { AtlasClient } from '@atlascompiler/sdk';
// Option A: Explicit manual API key
const client = new AtlasClient({
apiKey: 'atlas_your_api_key_here',
});
// Option B: Automatically loaded from process.env.ATLAS_API_KEY
// const client = new AtlasClient();
// Compile any web page into clean, structured Markdown
const response = await client.compile({
url: 'https://docs.github.com/en/rest',
});
if (response.status === 200) {
console.log('Title:', response.value.metadata.title);
console.log('Markdown:\n', response.value.markdown);
}Python
Install the official Python client library:
pip install atlascompilerfrom atlascompiler import AtlasClient
# Option A: Explicit manual API key
client = AtlasClient("atlas_your_api_key_here")
# Option B: Automatically loaded from os.environ["ATLAS_API_KEY"]
# client = AtlasClient()
response = client.compile({"url": "https://example.com"})
if response.status == 200:
print("Title:", response.metadata.get("title"))
print("Markdown:\n", response.markdown)Async support is built-in via AsyncAtlasClient.
Go
Install the official Go module:
go get go.atlas-compiler.com/sdkpackage main
import (
"context"
"fmt"
"log"
atlas "go.atlas-compiler.com/sdk"
)
func main() {
// Option A: Explicit manual API key
client := atlas.New("atlas_your_api_key_here")
// Option B: Automatically loaded from os.Getenv("ATLAS_API_KEY")
// client := atlas.New("")
res, _, err := client.Compile(context.Background(), atlas.CompileUrlRequest{
Url: "https://example.com",
})
if err != nil {
log.Fatal(err)
}
if res.Response200 != nil {
fmt.Println("Markdown:\n", res.Response200.Markdown)
}
}🛠 Supported Operations
Atlas exposes 5 core operations via REST, SDKs, and MCP:
Synchronous Operations (Real-Time)
scrape(POST /v1/scrape): Real-time Markdown, raw text, and link extraction for interactive agent loops (sub-1.5s).compile(POST /v1/compile): Deep 12-pass AST compilation with structural content graph extraction, Table of Contents, and link resolution.map(POST /v1/map): Rapid topological spider that crawls sitemaps and discovers reachable URLs across a domain without fetching bodies.
Asynchronous Operations (Queued Jobs)
crawl(POST /v1/crawls): Bounded, asynchronous multi-page website crawls with path glob filtering and progress tracking (202 Accepted).batch(POST /v1/batches): High-concurrency parallel retrieval for up to 500 URLs per job with unified status webhooks (202 Accepted).
💬 Community & Support
Discord: Join our active developer community at discord.gg/z9ktRhuudq for help, agent prompts, and feature discussions.
GitHub Discussions: Share recipes, ask questions, or propose ideas at Discussions.
Bug Reports: Open an issue using our Bug Report Template.
Security Inquiries: Email
support@atlas-compiler.comfor private disclosure.
📜 Contributing & License
Contributions are welcome! Please read our Contributing Guide and Code of Conduct.
Licensed under the MIT License.
This server cannot be deployed
Maintenance
Related MCP Connectors
Web pages as clean Markdown and JSON for AI agents, in fewer tokens, with an Evidence Record.
Web scraping for agents. Point it at a URL and it returns the page as clean markdown, JavaScript-rendered pages included. Point it at a site and it maps the URLs or crawls the section you need in the background, a few pages at a time so results fit in the conversation. Search the web and read full pages, extract fields with a JSON schema you define (validated, never invented), read a store's catalogue or a blog's posts from the platform's own feed, and check whether a page has changed. Failed requests cost nothing. The free plan includes 1,500 credits a month.
Web scraping for AI agents: scrape, search, crawl, map any website to markdown + JSON. No browser.
Web data for agents. Extract web pages, crawl websites, discover URLs, and search the web.
161
Related MCP Servers
- AlicenseAqualityAmaintenanceWeb content extraction for AI agents. 10 tools: scrape, crawl, map, batch, extract, summarize, diff, brand, search, research. Uses TLS fingerprinting to bypass anti-bot without a headless browser. Outputs LLM-optimized markdown with 67% fewer tokens than raw HTML.102,390AGPL 3.0
- AlicenseAqualityCmaintenanceConverts raw HTML into structured, AI-readable page maps with 97% token reduction, enabling agents to read, click, type, and navigate any web page.21336AGPL 3.0
- AlicenseBqualityAmaintenanceWeb scraping for AI agents: scrape, map, crawl and batch public pages into Markdown, tables and JSON, with an Evidence Record on every page. Blocked, empty and wrong pages come back as such, with the reason.392AGPL 3.0
- FlicenseNot gradedqualityAmaintenanceEnables AI agents to search multiple web providers in parallel and crawl single pages or entire sites into clean, deduplicated Markdown or JSON, including authenticated crawls.-