Maxun is an open-source web data platform that turns websites into structured APIs. Operations teams, growth analysts, and developers use it for recurring extracts from sites that change, gate content behind logins, or render in JavaScript. Instead of writing a scraper, a user records their own browsing once, and Maxun replays those steps as a scheduled robot delivering rows, files, or API responses.[1][2]
Quick verdict
Small operations and growth teams without scraping engineers gain the most from Maxun, because its recorder converts a single browsed session into a repeatable extraction robot. The concrete capability behind that recommendation is the combination of five robot types with hourly or daily schedules, webhooks, and API delivery. The material constraint is anti-bot depth and developer ergonomics: advanced stealth, captcha bypass, and pre-trained robots live only in the paid cloud, and self-hosters bring their own proxies. If the team wants no-code scheduling with a self-host option, start with Maxun; if it wants a code-first scraping API, evaluate Firecrawl instead.[1][2][3]
Capabilities
Record-and-run robots Demonstrate clicks, form fills, pagination, and logins in the browser, and Maxun stores the session as a robot that reruns on a schedule without rewritten code.[1][2]
Full-site crawling Point a crawl robot at a domain to discover pages with scoped link following, capturing each page's content instead of hand-maintained URL lists.[1][2]
Programmatic web search Run searches as a robot step with time-based filtering, receiving results as metadata or full page content for monitoring and research jobs.[1][2]
Scheduled change monitoring Set robots to run hourly or daily so price, stock, listing, or content changes surface automatically, with notifications on the managed cloud.[1][2]
Natural-language extraction Describe the wanted fields in plain language and let the AI-assisted mode navigate and pull them, which suits pages whose layout resists fixed selectors.[1]
Agent-ready interfaces The same engine answers the visual recorder, the REST API, SDKs, the CLI, and MCP clients, so a workflow defined by an analyst can later be driven by an agent.[1][2]
Login and document handling Recorder-based sessions carry authenticated state for pages behind logins, and document robots parse PDF, CSV, spreadsheet, Word, and image inputs into structured data.[1][2]
Credit billing for successful runs Cloud plans charge credits per page, row batch, screenshot, or search result, and only successful runs consume credit, which makes trial costs predictable.[1]
Who it fits
Analyst-led teams Teams with no scraping engineers cover recurring extracts through the recorder plus scheduling, work that would otherwise wait on developer time.[1][2]
Market and price trackers Hourly or daily monitors with notifications match catalogues, listings, and competitor pages that change often.[1][2]
Self-hosting shops The Docker deployment plus bring-your-own-proxy model fits teams that already run infrastructure and want data to stay on their boxes.[1][2]
Boundaries to check
Cloud-only anti-bot depth Proxy rotation, captcha bypass, and advanced stealth are cloud-only, so scrapers facing hardened targets have less to work with on a self-host.[1]
Recorder-first ergonomics Developers who live in code may prefer a scraper designed around SDK calls over defining work through a visual recorder first.[1][2]
Undocumented compliance surface No data-region statement or SSO documentation was found in the researched pages, so regulated procurement would need answers the public docs do not give.[1][2]
How it compares
Shared workflow Both Maxun and Firecrawl turn target sites into markdown, JSON, or datasets through scrape, crawl, and search operations, so a team comparing them evaluates the same adoption workflow. This comparison is editorial and not vendor-endorsed.[1][2][3]
Maxun edge Maxun leads on visual robot definition, scheduled monitoring out of the box, and a self-hostable core under AGPL-3.0, which suits analyst-led teams and infrastructure owners.[1][2][3]
Firecrawl edge Firecrawl leads as a developer-first API with managed infrastructure from the first call, which suits code-driven teams that would rather call endpoints than record browser sessions.[1][2]
Questions worth asking
What is the best Maxun alternative for non-technical users?
Is Maxun really no-code?
How much does Maxun cost per month?
As checked on October 11, 2026, Maxun Cloud lists Starter at $19, Growth at $99, and Pro at $399 per month in USD, plus a custom Premium tier, with a 7-day free trial and no card required. Billing runs on credits for successful runs, for example 3,500 credits and pages on Starter, while self-hosting the open-source core carries no licence fee beyond the operator's own infrastructure.[1]
Can you self-host Maxun?
Sources and editorial review
Source research. Reviewed on October 11, 2026. Written by Sanjib Sah. AI assisted; reviewed by a human editor.
We have not run or tested this product.
- Maxun - Open-Source Web Scraping and Data Extraction for AI
official · Observed October 11, 2026
Official homepage: open-source platform to scrape, crawl, search, monitor, and extract web data; no-code recorder, scheduled monitoring, REST API/webhooks/SDKs, stealth with proxy rotation, natural-language AI extraction, MCP/CLI interfaces, cloud signup and GitHub link.
- Maxun Pricing - plans from $19 per month
official · Observed October 11, 2026
Official pricing in USD: Starter $19/mo 3,500 credits, Growth $99/mo 100,000, Pro $399/mo 350,000, Premium custom; 7-day free trial, pay per successful run, credit rules per robot type, API/webhooks/schedules on all plans, SLA only on Premium.
- Maxun Documentation
official · Observed October 11, 2026
Official docs hub: extraction, crawling, scraping, search, monitoring, document parsing; no-code dashboard plus API, SDK, CLI, MCP; self-host section with Docker and local guides; cloud-vs-oss comparison; quickstart and robot overviews.
- Install Maxun with Docker Compose
official · Observed October 11, 2026
Official Docker Compose self-host guide: Docker Desktop prerequisite, generated JWT/session/encryption secrets, env file from ENVEXAMPLE, compose file from repo, frontend at PUBLIC_URL and backend at BACKEND_URL.
- Maxun Cloud versus Self-Hosted Open Source
official · Observed October 11, 2026
Official cloud-vs-oss matrix: core extraction/scraping/crawling/search/monitoring plus API/SDK/CLI/MCP on both; cloud-only deep extraction, managed stealth, proxy rotation, captcha bypass, auto robots, teams, managed notifications; self-host uses manual recorder login and own proxies.
- getmaxun/maxun on GitHub
official · Observed October 11, 2026
GitHub getmaxun/maxun metadata: TypeScript, AGPL-3.0, homepage maxun.dev, non-fork, non-archived, default branch develop, about 17.8k stars at observation; topics include playwright, web-scraper, self-hosted, no-code, rpa, browser-automation.
- Maxun LICENSE - AGPL-3.0
official · Observed October 11, 2026
Repository LICENSE file is the verbatim GNU Affero General Public License version 3 text, matching the API-reported AGPL-3.0 SPDX identifier.
- Firecrawl - Web Data API for AI Agents
official · Observed October 11, 2026
Official Firecrawl homepage: open-source web data API with search, scrape, map, crawl endpoints returning JSON with markdown; Python/Node SDKs; self-hostable core with managed cloud. Used only to evidence the substitutable scrape/crawl/search workflow shared with Maxun.
Compared with these products
No explicit proprietary baseline is recorded.