Citation Intelligence Mcp
@AutomateLab-tech
About Citation Intelligence Mcp
Free MCP server that tells your agent what LLMs cite, across Perplexity, Claude, ChatGPT, Gemini, Google AI Overviews, and Bing. One server, 6 AI citation sources, all via standard MCP tools. No API key required — open source and self-hostable.
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"citation-intelligence": {
"command": "npx",
"args": [
"-y",
"@automatelab/citation-intelligence"
]
}
}
}Tools
No tools detected
Fetch the live tool list by running this server in a temporary sandbox using the button above.
Overview
What is Citation Intelligence MCP?
A free, self-hosted MCP server that tells your agent which URLs get cited by AI search engines—across Perplexity, Google AI Overviews, ChatGPT, Claude, Gemini, and Bing. It’s for developers, SEOs, and content creators who need programmatic citation data without a hosted dashboard.
How to use Citation Intelligence MCP?
Install via npx -y @automatelab/citation-intelligence (requires Node 20+). Set environment variables for desired API keys (Perplexity, SerpAPI, Anthropic, OpenAI, Gemini, Bing, Brave). Use any MCP-compatible client (e.g., Claude Desktop, Cursor, Claude Code) with stdio transport.
Key features of Citation Intelligence MCP?
- Query cross-engine citation consensus matrix
- Domain-level citation check with per-engine breakdown
- Predict citation likelihood without firing an LLM
- Trend and freshness analysis for citations
- AI crawler access and schema audit tools
- Local caching with configurable TTL
Use cases of Citation Intelligence MCP?
- Check if your own domain is cited by AI search engines for target queries
- Identify competing domains cited across multiple engines for a query
- Monitor citation trends over time for editorial or GEO planning
- Audit your site’s AI crawler accessibility and structured data health
- Programmatically compare citation visibility between any set of URLs
FAQ from Citation Intelligence MCP
What does the surface field mean in responses?
It tells you how the data was collected: consumer_scrape (real UI proxy), api_proxy (API call, directional), web_rank (traditional SERP), or static_signal (offline). Always check this before drawing conclusions.
Do I need paid API keys to use the server?
No. Several engines have free tiers: Perplexity (free Sonar on signup), SerpAPI (100 free Google AI Overview lookups/month), Gemini (free tier), Bing (free), Brave (2000/month). Only Claude and ChatGPT require paid keys.
Where is my citation data stored?
All data stays on your machine. API calls go directly from your computer to the vendor. A local cache file lives at ~/.config/citation-intelligence/cache.json. No data is uploaded to any third-party backend.
Can I run Citation Intelligence MCP without any API keys?
Yes. The citations.predict tool scores citation likelihood from public signals (Wikipedia, schema.org, llms.txt, GitHub) and requires zero keys. Other tools become available as you set each key.
Is this server compatible with Claude Desktop?
Yes. Add it to the Claude Desktop config via stdio transport using the npx command and your chosen environment variables. Any MCP client that supports stdio works the same way.
Frequently asked questions
What does the `surface` field mean in responses?
It tells you how the data was collected: `consumer_scrape` (real UI proxy), `api_proxy` (API call, directional), `web_rank` (traditional SERP), or `static_signal` (offline). Always check this before drawing conclusions.
Do I need paid API keys to use the server?
No. Several engines have free tiers: Perplexity (free Sonar on signup), SerpAPI (100 free Google AI Overview lookups/month), Gemini (free tier), Bing (free), Brave (2000/month). Only Claude and ChatGPT require paid keys.
Where is my citation data stored?
All data stays on your machine. API calls go directly from your computer to the vendor. A local cache file lives at `~/.config/citation-intelligence/cache.json`. No data is uploaded to any third-party backend.
Can I run Citation Intelligence MCP without any API keys?
Yes. The `citations.predict` tool scores citation likelihood from public signals (Wikipedia, schema.org, llms.txt, GitHub) and requires zero keys. Other tools become available as you set each key.
Is this server compatible with Claude Desktop?
Yes. Add it to the Claude Desktop config via stdio transport using the `npx` command and your chosen environment variables. Any MCP client that supports stdio works the same way.
Basic information
More AI & Agents MCP servers
21st.dev Magic AI Agent
21st-devIt's like v0 but in your Cursor/WindSurf/Cline. 21st dev Magic MCP server for working with your frontend like Magic
UnificAlly MCP
unifically-devUnificAlly MCP server. One API for 100+ AI video, image, music and speech models. https://mcp.unifically.com/mcp
Simplepages
SimplepagesBuild and edit landing pages and websites from your AI assistant, and pull how they are performing.
Orders of Magnitude — x402 API Catalog
OrdersOfMagnitudeLLC1000+ pay-per-call API endpoints across finance, weather, geography, economic data, agent infrastructure (memory, coordination, secrets, simulation, verification), and NS compute. Payments via x402 protocol on Base/USDC.
XDC AI
XDC AI is a remote MCP server available at https://xdcai.tech.
Comments