M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
M

Maxun MCP Server

Official
Turn Any Website Into A Structured API
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #348 of 438
Stars
★ 17.8k Very popular
Transport
stdio (local process)
Runtime
Docker · Node.js · Hosted (no local runtime)
Credentials
No credential needed
License
AGPL-3.0
Last commit
Tools
9
41FMRS · D

Maxun offers a feature-rich no-code data extraction platform with integrated MCP support. Highlights include ease of use and flexibility, allowing creation of robots via recording or AI description, and the MCP server exposes these capabilities to AI assistants. However, users should be aware of scraping legal and compliance risks, and potential stability issues with dynamic websites. Overall, it's a promising tool but still in early stages.

Strongest · Maintenance 12/20 Weakest · Reliability 5/20

Reliability
5/20
Security and permissions
6/20
Maintenance
12/20
Documentation
8/20
Setup experience
10/20
Why each score
Reliability 5/20
The project is a full web application rather than a standalone MCP server. MCP support is only mentioned in the README and possibly in code snippets; no server manifest or dedicated MCP entry point is present. Static review cannot verify MCP handshake or tool behavior, and no MCP configuration examples are provided. Thus, reliability is capped at 5.
Security and permissions 6/20
The app is a self-hosted platform for data extraction and may make external network requests. Source code not fully reviewed, but no obvious malicious behavior found. Lacks clear permission model and confirmation for dangerous operations; security documentation is thin. Thus, score is capped at 6.
Maintenance 12/20
Repository is active with many stars and recent commits, not archived, and licensed under AGPL-3.0. Open issues exist, but no release versioning or security response channel is evident. Therefore, score is 12.
Documentation 8/20
README is comprehensive with installation, usage, and feature descriptions, and links to detailed documentation. However, as an MCP server, it lacks MCP-specific usage instructions, tool parameters, and troubleshooting. Thus, score is 8.
Setup experience 10/20
Installation via Docker and local setup is documented with clear steps and environment variable docs. But as an MCP server, no configuration example for MCP clients is provided, and the full app must be set up. Hence, score is 10.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkControls a browser

Best for

  • Non-technical users who want to extract data from websites without coding.
  • Developers looking to turn websites into structured APIs for integration.
  • Individuals and businesses needing automated, scheduled scraping.
  • Teams that want to self-host their scraping infrastructure.

Not for

  • Large-scale scraping requiring advanced anti-bot evasion (e.g., complex CAPTCHAs) beyond simulated browsing.
  • Ultra-real-time scraping of extremely fast-changing sites.
  • Production environments needing commercial support or SLAs.
  • Users who only need simple single-page HTML scraping without full robot capabilities.

Required permissions

  • Network access: connects to target websites to open pages, click, fill forms, etc., so it requires internet access.
  • API key: needs an API key from the Maxun instance for authentication.
  • Data access: may read and extract website content, including behind-login data if configured.
  • Storage and scheduling: may need permissions to schedule tasks and store extracted data in some configurations.

Risks and side effects

  • Legal compliance: scraping may violate website terms or applicable laws, especially with personal data; ensure you have legal rights.
  • IP bans: scraping can trigger anti-bot measures, leading to temporary or permanent IP bans.
  • Data accuracy: dynamic sites or layout changes may cause extraction failures or inaccurate data.
  • Security: when self-hosting, protect infrastructure and avoid API key leaks; data stored locally requires secure management.

Setup

Before you start

Runtime：Docker · Node.js · Hosted (no local runtime)

  1. Ensure you have a running Maxun instance (either the hosted app.maxun.dev or a self-hosted Docker/local installation).
  2. Install the MCP server package via npm or pip (see official docs or package registries for exact commands).
  3. Configure environment variables like MAXUN_API_KEY and MAXUN_INSTANCE_URL (see docs).
  4. Add the MCP server to your client config: e.g., in Claude Desktop's config file, add a JSON block specifying the command and arguments.
  5. Restart the client and you should be able to use Maxun MCP tools.

Check that it works

Check that tools like extract, scrape, crawl, search, and get_robots appear in the client's tool list; calling get_robots and getting a (possibly empty) robot list confirms the connection.

Troubleshooting

  1. Connection failures: check that the Maxun instance URL is accessible and API key is correct.
  2. Tool errors: ensure the robot has been created in Maxun beforehand and parameters are correct.
  3. Extraction failures: try re-recording the robot in Maxun, or check if the target site changed layout.
  4. Scheduling issues: verify timezone and schedule configuration, and check Maxun logs.

Things to try

Once connected, you can ask your AI assistant things like:

  • Extract 10 property listings from Airbnb
  • Convert the full example.com page to Markdown and take a screenshot
  • Crawl example.com and extract content from every page
  • Search the web for AI news from the past week

Tools 9

extract read-only
Extract structured data from a webpage (supports modes: recorder, AI).
scrape read-only
Convert the entire webpage into clean Markdown or HTML and capture screenshots.
crawl read-only
Crawl entire websites and extract content from every relevant page, with control over scope and discovery.
search read-only
Run automated web searches to discover or scrape results, with support for time-based filters.
schedule_run writes
Schedule robot runs for time-based automation.
get_robot_run read-only
Get detailed information about a specific robot run.
get_robot_runs read-only
List the run history for a robot.
get_robot read-only
Get a specific robot's configuration.
Show 1 more tools
get_robots read-only
List all available robots.

Use cases

Market research: gather competitor data, prices, and reviews.
Lead generation: extract contact details from directories.
Content aggregation: scrape articles from multiple sources.
Data monitoring: track changes on websites over time.
AI applications: prepare clean text for RAG or LLMs.

Supported clients

Claude Desktop

Listed from the project's documentation, not tested by this site.

Overview

Maxun is an open-source no-code web data platform for extracting, scraping, crawling, and searching structured data from real websites. Combining Playwright browser automation with optional AI (LLM) power, it offers Recorder and AI modes to create reusable robots without programming. You can access data via RESTful API, SDK, or CLI, and schedule runs, handle pagination, and extract behind login. The project is AGPL-3.0 licensed and self-hostable.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision 1ec47a5aa8ab Data synced 2026-10-12 Read the FMRS scoring method