Seo Crawler Mcp
Run agent-driven site crawls that surface SEO issues on your product or marketing site and persist results for fix lists.
Overview
SEO Crawler MCP is a MCP server for the Launch phase that crawls and analyses websites for SEO errors using Crawlee with SQLite storage.
What is this MCP server?
- Crawls and analyses websites for SEO errors using Crawlee (v2.0.1 server)
- Persists crawl output with SQLite-backed storage under a configurable OUTPUT_DIR
- Stdio MCP via npm @houtini/seo-crawler-mcp
- Optional DEBUG=true for verbose logging during crawl troubleshooting
- GitHub repo houtini-ai/seo-crawler-mcp with MCP server.schema.json metadata
- Server version 2.0.1
- Crawlee-based crawl with SQLite storage
- 1 required environment variable (OUTPUT_DIR); optional DEBUG flag
Community signal: 16 GitHub stars.
What problem does it solve?
Shipping a solo-built site without a structured crawl leaves broken SEO issues invisible until traffic flatlines or Search Console alerts pile up.
Who is it for?
Builders auditing their own marketing site, docs domain, or staging URL from the IDE with repeatable MCP-driven crawls.
Skip if: People who only need Knowledge Graph entity facts or who cannot allocate a local output directory for crawl artifacts.
What do I get? / Deliverables
Crawl results and SQLite-backed reports land in OUTPUT_DIR so your agent can prioritize fixes before and after you go live.
- Crawl-derived SEO issue signals from your site
- Persisted crawl session data under OUTPUT_DIR
- Agent-readable output for remediation backlogs
Recommended MCP Servers
Journey fit
Launch is the canonical shelf because technical and on-page SEO audits matter most when you are preparing to be indexed and discovered. SEO subphase matches Crawlee-based crawling and error analysis aimed at findability, not core application feature work.
How it compares
On-site SEO crawl MCP with Crawlee storage, not a SERP rank tracker or Google Knowledge Graph entity lookup.
Common Questions / FAQ
Who is SEO Crawler MCP for?
Solo builders using MCP hosts who want crawl-based SEO error detection on sites they control, without a separate crawler UI.
When should I use SEO Crawler MCP?
Use it during launch SEO prep or after shipping when you need a fresh technical crawl before indexing pushes or content sprints.
How do I add SEO Crawler MCP to my agent?
Register @houtini/seo-crawler-mcp as a stdio MCP server, set OUTPUT_DIR to a writable folder, and optionally set DEBUG=true.