houtini-ai avatar

Fanout

houtini-ai/fanout-mcp

Fan-Out analyzes web pages to figure out which user queries your content actually answers and which ones it misses. It scrapes a URL, then generates 15-25 query variations using Google's patented fan-out methodology (equivalent phrasings, specs, comparisons, follow-ups, clarifications). For each variant, it uses Self-RAG to check if your content covers it and pulls exact quotes as evidence. You get a coverage score, gap-by-gap recommendations, and a visual dashboard. The hybrid mode is the sweet spot: point it at your article with a target keyword and it shows you exactly what's missing for GEO and AI search citation. Built on Anthropic's API, runs via stdio, outputs interactive markdown reports.

GitHub stars
14
tool
1
transport
STDIO
last updated
Oct 24, 2026

Tools

1 tools

Public tool metadata for what this MCP can expose to an agent.

analyze_content_gapAnalyzes a URL for content gaps, with optional keyword fan-out mode that generates query variants.

Analyzes a URL for content gaps, with optional keyword fan-out mode that generates query variants.

No parameter schema in public metadata yet.

Sponsors

Readme

Fan-Out MCP

Socket Badge

npm version MCP Registry License TypeScript MCP Research Backed

Advanced content gap analysis for the AI search era

Analyze your content to discover what user queries it covers (and misses) using the same techniques AI search engines use internally.


View the full README on GitHub

Details

Repository
github.com/houtini-ai/fanout-mcp
Website
houtini.com/query-fan-out-mcp-for-ai-search-optimisation
Package
@houtini/fanout-mcp
Transport
STDIO
Registry
MCP registry, active

More from houtini-ai