Scrawly MCP Server

What Scrawly MCP Server does

Scrawly MCP Server MCP server gives AI assistants access to audits produced by the Scrawly desktop SEO crawler. It is intended for working with local audit data, including the findings, affected pages, explanations, and suggested remediation associated with a crawl.

Scrawly itself crawls websites and evaluates 250 checks across 23 areas. These include crawlability, indexability, response codes, redirects, metadata, headings, canonicals, content quality, structured data, internal linking, performance, AI-search readiness, agentic-web access, images, hreflang, sitemaps, security, mobile behavior, JavaScript rendering, pagination, and WordPress-specific issues.

The MCP interface is useful after an audit has been run: an AI assistant can inspect issues and help determine what should be addressed. For WordPress sites, the project also supports safe and reversible fixes; proposed changes are confirmation-gated rather than applied without approval.

How it works

Scrawly is a desktop application for macOS, Windows, and Linux. It stores audits locally, and the MCP server lets AI assistants work with those local audits. The available description supports two broad operations: querying audit results and proposing fixes for confirmation.

Audit findings include the reason an issue matters, how it can be fixed, and the exact pages affected. The underlying crawler can use HTTP inspection and headless Chromium rendering, and it can produce saved crawl profiles, site graphs, crawl comparisons, scheduled audits, and reports. The MCP server is described as an interface for assistants to work with audits; the provided material does not specify individual MCP method names or a transport.

Setup and configuration

Install Scrawly using the project’s installer for the relevant operating system. macOS and Linux users can run the published shell installer, while Windows users can use the PowerShell installer. A manual installation requires Python 3.11 or newer and git; the installer creates a private Python environment and downloads the crawler browser.

After installation, open Scrawly as a desktop application and run or load an audit. The supplied material does not document MCP-specific configuration files, command arguments, environment variables, or client setup steps, so those details should be checked in the project’s MCP documentation before deployment.

Tools and capabilities

Scrawly MCP Server MCP server supports:

• Querying local Scrawly audit data.
• Reviewing findings and the pages affected by them.
• Working with explanations and recommended remediation.
• Proposing fixes that require confirmation.
• Supporting the WordPress fix workflow described by Scrawly.

The crawler behind the interface also covers technical SEO, GEO, and AI-search checks, including AI crawler access, llms.txt, entity and structured-data signals, and citation readiness. Those audit categories are capabilities of Scrawly; the excerpt does not provide a separate MCP tool list for each category.

Limitations and notes

The MCP server works with audits from Scrawly rather than replacing the crawler’s desktop workflow. The provided material does not state whether it can start crawls, access remote audits, or operate without the Scrawly application.

Fixes are described as safe and reversible for WordPress, but they are confirmation-gated. An assistant should therefore propose a change and wait for approval instead of assuming that a recommendation has been applied.

Scrawly is released under the MIT License. It is presented as a free, open-source application with no account requirement and no page cap, but those statements describe Scrawly generally and do not establish additional MCP-server hosting or access guarantees.