Skip to main content
Glama
Sign In
siteone-mcp-server

siteone-mcp-server

by Dorkside
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

siteone-mcp-server

MCP server that wraps the SiteOne Crawler CLI for SEO auditing and site analysis. Use it with Claude Code, Claude Desktop, or any MCP-compatible client.

Install SiteOne Crawler

The auto-installer downloads the correct SiteOne Crawler binary for your platform:

npx -y siteone-mcp-server --install

This installs to ~/.siteone-crawler/ and the MCP server will auto-detect it — no further configuration needed.

Supports macOS (arm64, x64) and Linux (arm64, x64). Windows users should download manually.

Related MCP server: seocli

Quick Start

Claude Code

claude mcp add siteone -- npx -y siteone-mcp-server

Claude Desktop

Add to your MCP settings (~/.claude/mcp_settings.json):

{
  "mcpServers": {
    "siteone": {
      "command": "npx",
      "args": ["-y", "siteone-mcp-server"]
    }
  }
}

Available Tools

Tool

Description

crawl_site

Full site crawl — returns JSON with SEO, performance, and security metrics for all URLs

crawl_single_page

Single page analysis — fast, lightweight audit of one URL

generate_sitemap

Crawl a site and generate an XML sitemap file

export_markdown

Crawl a site and export all pages as markdown files

get_crawl_summary

Quick shallow crawl for health check statistics

Configuration

The server resolves the SiteOne binary in this order:

  1. --siteone-bin CLI argument (highest priority)

  2. SITEONE_BIN environment variable

  3. ~/.siteone-crawler/crawler (auto-installed location)

  4. crawler in PATH (fallback)

CLI argument

claude mcp add siteone -- npx -y siteone-mcp-server --siteone-bin=/path/to/crawler

Or in Claude Desktop config:

{
  "mcpServers": {
    "siteone": {
      "command": "npx",
      "args": ["-y", "siteone-mcp-server", "--siteone-bin=/path/to/crawler"]
    }
  }
}

Environment variables

Variable

Purpose

Default

SITEONE_BIN

Path to the SiteOne crawler binary

Auto-detected (see above)

SITEONE_OUTPUT_DIR

Working directory for crawl outputs

Current working directory

Examples

Once configured, ask Claude:

Development

# Install dependencies
npm install

# Build
npm run build

# Watch mode
npm run dev

# Test with MCP Inspector
npm run inspect

License

MIT

Available Tools

5 tools
crawl_single_pageA

Analyze a single page with SiteOne Crawler. Returns JSON data for just the specified URL. Fast and lightweight — use for quick page-level audits.

Inputs: url (required), device, extra_columns.

crawl_siteA

Run a full site crawl with SiteOne Crawler. Returns structured JSON with all crawled URLs and their SEO, performance, and security metrics. Use for comprehensive site audits.

Inputs: url (required), device, timeout, workers, +6 more.

export_markdownA

Crawl a website and export each page as a markdown file. Returns the export directory path and a crawl summary. Use for content migration, offline analysis, or feeding pages into other tools.

Inputs: url (required), max_depth, output_dir.

generate_sitemapA

Crawl a website and generate an XML sitemap file. Returns the sitemap file path and a crawl summary. Use for SEO audits or submitting sitemaps to search engines.

Inputs: url (required), max_depth, output_file.

get_crawl_summaryA

Run a quick, shallow crawl and return only high-level statistics: status code distribution, response times, and error counts. Best for fast site health checks before running a full crawl.

Inputs: url (required), max_depth, rows_limit.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A4.2/5.0

    Scored across 5 tools

    Disambiguation5/5

    Each tool has a distinct purpose: full site crawl, single page crawl, sitemap generation, markdown export, and summary statistics. Even the overlapping crawl operations differ clearly in scope and output. An agent can reliably select the right tool based on whether it needs metrics, a sitemap, markdown files, or a quick health check.

    Naming Consistency5/5

    All tool names follow a consistent verb-first snake_case pattern: crawl_site, crawl_single_page, generate_sitemap, export_markdown, get_crawl_summary. The verbs are descriptive and the noun targets are clear. This is a textbook example of predictable naming.

    Tool Count5/5

    Five tools is well-scoped for a website crawling server. Each tool covers a distinct use case without redundancy, and the count is neither too sparse nor bloated. The set feels complete for the domain.

    Completeness5/5

    The tool surface covers the core operations expected from a crawler: full audit, single page check, sitemap generation, content export, and summary stats. There are no obvious gaps that would block an agent from achieving typical crawl-related tasks. Combined with the provided descriptions, the set appears comprehensive.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Resources

    Related MCP Connectors

    Related MCP Servers

    • crawlie-mcp

      spronta
      A
      license
      Not graded
      quality
      C
      maintenance
      Enables LLM agents to crawl and audit websites for technical SEO and GEO issues, providing actionable fixes via tools like crawl_site and explain_issue.
      107 npm
      119
      MIT

    • seocli

      lennney
      A
      license
      Not graded
      quality
      D
      maintenance
      Enables AI agents to crawl and audit websites for SEO issues, returning structured JSON reports with errors, warnings, and key statistics.
      MIT

    • seo-crawler-mcp

      hna2810
      F
      license
      A
      quality
      B
      maintenance
      Enables comprehensive website SEO analysis including crawling, on-page audits, site structure visualization, and content topic classification with taxonomy mapping.
      7
      1
      -