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

CrawlSEO MCP Server

Community
Open-source SEO monitoring: Google Search Console + Site Crawler + Core Web Vitals + MCP Server in one self-hosted dashboard.
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #158 of 438
Stars
★ 630 Very popular
Transport
stdio (local process)
Runtime
Node.js 20+ · Docker
Credentials
API key / credential required
License
MIT
Last commit
Tools
10

Static review · not runListed 2026-08-07

FMRS 57/100 · C · Weakest Reliability 8/20 — Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkConnects to a database

Best for

  • Founders, indie developers, and small teams
  • Users who want self-hosted SEO tools with data privacy
  • Users who want to leverage AI assistants for quick SEO insights

Not for

  • Users needing advanced enterprise-grade SEO features (e.g., large-scale keyword research, paid data)
  • Non-technical users, as self-hosting and configuration require technical skills
  • Users requiring official support (this is an open-source community project)

Required permissions

  • Requires access to the local server's database (PostgreSQL)
  • Requires Google OAuth credentials to access Google Search Console data
  • Requires network access to run crawls and fetch Core Web Vitals (e.g., via PageSpeed Insights)

Risks and side effects

  • Data security risk: if the self-hosted instance is exposed to the internet without proper protection, data leakage may occur.
  • Dependence on third-party services (Google, PageSpeed Insights, DataForSEO) for availability and quota limits.
  • Open-source software may have unknown vulnerabilities; you must maintain and update it yourself.
  • Misconfiguration can prevent it from working, such as missing environment variables or a down database.

Setup

Before you start

Runtime：Node.js 20+ · Docker

DATABASE_URL requiredsecret PostgreSQL connection string pointing to your database instance.
NEXTAUTH_SECRET requiredsecret Session encryption key, generated with openssl rand -hex 32.
GOOGLE_CLIENT_ID required Google OAuth client ID, created in the Google Cloud Console credentials page.
GOOGLE_CLIENT_SECRET requiredsecret Google OAuth client secret, created together with the client ID.
Other optional settings (2)
NEXTAUTH_URL optional Base URL of the app; auto-detected in most environments.
CRAWLSEO_IMAGE optional Optional; overrides the Docker image tag used by Compose (e.g. pinned version or locally built image).
  1. Clone the repo: git clone https://github.com/crawlseo/crawlseo.git and enter the directory.
  2. Install dependencies: npm install.
  3. Configure environment: copy .env.example to .env.local and set DATABASE_URL, NEXTAUTH_SECRET, GOOGLE_CLIENT_ID, and GOOGLE_CLIENT_SECRET.
  4. Start the database (e.g., with Docker): docker compose up -d db.
  5. Run migrations: npx prisma migrate dev --name init.
  6. Start the app: npm run dev.
  7. Configure the MCP server in a supported client (e.g., Claude Desktop) by adding the install_config JSON.
claude_desktop_config.json 
{
  "mcpServers": {
    "crawlseo": {
      "command": "npx",
      "args": [
        "tsx",
        "mcp/server.ts"
      ],
      "cwd": "/path/to/crawlseo"
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "crawlseo": {
      "command": "npx",
      "args": [
        "tsx",
        "mcp/server.ts"
      ],
      "cwd": "/path/to/crawlseo"
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add crawlseo -- npx tsx mcp/server.ts

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

The client's tool list should show the 10 crawlseo tools (e.g. list_sites, run_crawl); calling list_sites and getting back your added sites confirms the connection works.

Troubleshooting

  1. Ensure `.env.local` is configured and all required environment variables are set.
  2. Ensure PostgreSQL is running and accessible, with `DATABASE_URL` correct.
  3. Run `npx prisma migrate dev` to initialize the database schema.
  4. Check app logs for errors; ensure all dependencies are installed.
  5. Check MCP configuration file path and command in your client.

Things to try

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

  • List all my sites in CrawlSEO
  • Run a crawl and tell me the biggest issues found
  • What are my top keywords for the last 28 days?
  • Show me striking-distance keyword opportunities

Tools 10

list_sites read-only
List all added sites.
get_site_overview read-only
Get overview data for a specific site.
get_keywords read-only
Get keyword rankings and performance data.
get_pages read-only
Get page performance data.
get_traffic read-only
Get traffic statistics.
run_crawl writes
Trigger a crawl of the site.
get_crawl_status read-only
Get the status of the most recent crawl.
get_crawl_issues read-only
Get issues found during the crawl.
Show 2 more tools
get_vitals read-only
Get Core Web Vitals (LCP, CLS, INP, etc.).
get_opportunities read-only
Get SEO opportunities (e.g., low CTR, content decay).

Use cases

Query SEO data in natural language from your terminal, without opening the dashboard.
Let AI agents automatically analyze keyword performance, page issues, and generate optimization suggestions.
Integrate SEO monitoring into your existing AI workflows for automated reporting.

Supported clients

Claude Desktop

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

Overview

The CrawlSEO MCP Server is part of the CrawlSEO platform, enabling AI agents (such as Claude Code, Claude Desktop, and Cursor) to query your SEO data directly. It provides 10 tools covering site listing, overview, keyword and page data, traffic metrics, crawl and audit, Core Web Vitals, and SEO opportunities. The server runs over stdio transport and requires a locally running CrawlSEO instance with proper environment configuration.

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 92c254dae84d Data synced 2026-10-12 Read the FMRS scoring method