sightvane-mcp
Enables AI assistants to perform comprehensive SEO and GEO measurements, including site audits, keyword research, ranking tracking, and brand visibility analysis across search engines and generative AI platforms.
README
sightvane-mcp
An MCP server that turns any AI assistant into an SEO + GEO measurement instrument.
Sightvane measures the two ways buyers find you now: classic search (technical site health, keywords, rankings, performance) and generative engines (whether ChatGPT, Claude, Perplexity, and Google AI Overviews mention your brand and cite your pages). This server exposes the full instrument as 16 Model Context Protocol tools, so an assistant like Claude can run complete SEO sessions: crawl a site, research keywords, sample AI-answer panels, and report visibility scores, in one conversation.
Built by Sightvane. New to GEO? Read the plain-language guide.
Why this exists
- Zero fixed cost. The crawler, keyword engine (live autocomplete graph), and assistant-driven GEO sampling need no API keys at all. Optional free/cheap keys unlock PageSpeed, domain authority, SERP snapshots, and multi-engine sampling.
- The assistant IS the answer engine. The claude-session loop lets the assistant answer realistic buyer prompts naturally, record its own answers, and measure the brand mentions inside them. One sampling engine, effectively free.
- Local-first. Your data and keys stay in a local JSON store. Nothing syncs anywhere.
Quickstart
git clone https://github.com/BaseHoss/sightvane-mcp.git
cd sightvane-mcp && npm install
Register with Claude Code:
claude mcp add sightvane -- node /path/to/sightvane-mcp/src/index.js
Or in any MCP client config:
{
"mcpServers": {
"sightvane": { "command": "node", "args": ["/path/to/sightvane-mcp/src/index.js"] }
}
}
Then ask your assistant: "run an SEO session on example.com".
The 16 tools
| Tool | What it does |
|---|---|
sightvane_list_projects / sightvane_create_project |
Manage measured domains (brand + competitors enable share of voice) |
sightvane_run_audit / sightvane_get_audit |
Polite, sitemap-seeded crawl; 30+ technical checks; health score across 9 categories |
sightvane_keyword_ideas |
Long-tail expansion from the live suggest graph with intent classification |
sightvane_track_keywords / sightvane_check_rankings |
Rank-tracking list + live SERP positions (serper.dev key) |
sightvane_pagespeed |
Lighthouse + Core Web Vitals via Google PageSpeed Insights |
sightvane_authority |
Domain authority for up to 100 domains (OpenPageRank, free key) |
sightvane_serp |
Live SERP snapshot: organic top 20, People Also Ask, related |
sightvane_geo_generate_panel / sightvane_geo_get_panel |
Funnel-staged buyer-question panels for AI-visibility sampling |
sightvane_geo_record |
Record an answer-engine response; extracts mentions, citations, position, sentiment |
sightvane_geo_sample |
Sample configured engine APIs (OpenAI, Anthropic, Perplexity, Gemini) |
sightvane_geo_report |
Visibility score, mention/citation rates, share of voice, per-stage and per-engine breakdowns |
sightvane_recommendations |
Prioritized action plan from the latest audit + visibility report |
The honest-sampling rule
GEO numbers are only meaningful if the sampling is unbiased. When an assistant answers panel
prompts for sightvane_geo_record, it must answer exactly as it would for any user, never
deliberately mentioning the measured brand. A flattering panel is a false map.
Optional keys (environment or the Sightvane app's Settings)
All engines degrade gracefully with zero keys. Optional: Google PageSpeed, OpenPageRank (free), serper.dev (free tier), OpenAI / Anthropic / Perplexity / Gemini for multi-engine GEO sampling.
Data lands in ./data by default; set SIGHTVANE_DATA to point elsewhere (for example, at
the Sightvane dashboard's data directory so results appear there live).
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.