data-ai MCP Server
The AI-native technical SEO crawler. Open-source MCP server for Claude / Cursor / Codex — 37 tools, 50+ checks, unlimited pages, WAF detection, ephemeral by design. Built on LibreCrawl. MIT.
Discovered via github-topic:mcp and last synced Today.
Install instructions not detected yet
Check the source repository for the latest setup steps.
Language variant detection for multilingual sites
**One-call full audit** — crawl + site checks + 30+ checks + report
List all saved crawls
Merge GSC indexing errors into any audit report
Missing alt text (per-page count + total), broken image references
Raw JSON export
What it catches
Internal authority map — top linked pages, orphans, dead ends, anchor text
Stop running crawl
Core Web Vitals for one URL (Google PSI API)
Start async crawl, returns `crawl_id`
Site link graph (nodes + edges)
`5080`
Screaming Frog Free
Schema.org / JSON-LD for one URL — rich result mapping
Pause a running crawl
unset
Tiered by subscription plan
Schema coverage across multiple URLs
Resume a paused crawl
`~/librecrawl-reports`
What it does
Show current crawler settings
Purpose
Instant: robots.txt, sitemap, HTTPS, www — no crawl needed
Re-generate report from a past crawl
Batch CWV for up to 25 URLs, ranked worst-first
Poll crawl progress
`5081`
Exclude false-positive patterns from results
`~/librecrawl-mcp`
Desktop crawler (Screaming Frog SEO Spider™)<sup>1</sup>
Hidden
❌
✅
Use
`http://127.0.0.1:5080`
`127.0.0.1`
`http`
`~/.librecrawl/upstream/users.db`
What's inside