TrueSource GEO MCP Server
Enables AI assistants to audit websites for AI-readiness, generate AI-optimized robots.txt and llms.txt files, and create VibeTags for emotional AI brand resonance.
README
TrueSource GEO MCP Server
AI-Readiness auditing as an MCP tool surface — for Claude Desktop, Cursor, VS Code Copilot, and 70+ other MCP clients.
What is this?
This is a Model Context Protocol (MCP) server that exposes TrueSource's GEO-Audit methodology as standardized tools. Any MCP-compatible AI assistant can:
- Audit websites for AI-readiness (robots.txt, llms.txt, schema markup, E-E-A-T signals)
- Generate GEO files (AI-optimized robots.txt, llms.txt)
- Create VibeTags for emotional AI brand resonance
- Check robots.txt for AI bot allow/block status
Quick Start
Option A: npx (Recommended)
⚠️ The
-yflag is critical — without it, npx silently waits for install confirmation in the background and the server freezes.
Claude Desktop
Edit ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"truesource": {
"command": "npx",
"args": ["-y", "truesource-geo-mcp"],
"env": {
"TRUESOURCE_API_KEY": "your-api-key-here"
}
}
}
}
Cursor / VS Code
Add to .cursor/mcp.json or VS Code MCP settings:
{
"mcpServers": {
"truesource": {
"command": "npx",
"args": ["-y", "truesource-geo-mcp"],
"env": {
"TRUESOURCE_API_KEY": "your-api-key-here"
}
}
}
}
Option B: Local Build
cd packages/geo-mcp
npm install
npm run build
{
"mcpServers": {
"truesource-geo": {
"command": "node",
"args": ["/absolute/path/to/packages/geo-mcp/build/index.js"],
"env": {
"TRUESOURCE_API_KEY": "your-api-key-here"
}
}
}
}
3. Restart Your MCP Host & Use
Ask your AI assistant:
- "What's the AI-readiness score for my-website.com?"
- "Generate a robots.txt for https://example.com"
- "Check which AI bots are blocked on competitor.com"
- "Create VibeTags for our brand at my-website.com"
Available Tools
| Tool | Trigger | Description |
|---|---|---|
geo_score |
"audit", "score", "check AI readiness" | Full AI-readiness audit → 0-100 score, grade, checks, recommendations |
geo_inject |
"generate robots.txt", "create llms.txt" | Generate AI-optimized robots.txt + llms.txt (ready to deploy) |
vibetags_generate |
"brand perception", "VibeTags", "how AI sees" | Emotional AI brand resonance analysis (4-layer VibeGap Bridge) |
geo_check_robots |
"robots.txt", "which bots blocked" | Quick robots.txt AI bot status check (9 bots) |
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
TRUESOURCE_API_KEY |
Yes | — | API key for Railway backend authentication |
TRUESOURCE_API_URL |
No | Railway production URL | GEO-Inject API base URL |
TRUESOURCE_TIMEOUT |
No | 30000 |
Request timeout in ms (MCP hosts timeout at ~60s) |
Testing with MCP Inspector
npm run inspect
This opens the interactive MCP Inspector at http://localhost:6274, where you can:
- See all registered tools and their schemas
- Invoke tools with test data
- Inspect JSON-RPC request/response pairs
Architecture
┌─────────────────────────────┐
│ MCP Host (Claude Desktop) │
│ User: "Audit example.com" │
└─────────────┬───────────────┘
│ Stdio (JSON-RPC 2.0)
▼
┌─────────────────────────────┐
│ truesource-geo-mcp v1.2.0 │
│ (TypeScript MCP Server) │
│ │
│ Tools: │
│ ├── geo_score │
│ ├── geo_inject │
│ ├── vibetags_generate │
│ └── geo_check_robots │
└─────────────┬───────────────┘
│ HTTPS + Bearer Auth
▼
┌─────────────────────────────┐
│ Railway API │
│ (FastAPI / Python) │
│ truesource-mcp-api │
│ │
│ POST /audit/robust │
│ POST /activate/robots │
│ POST /activate/llms │
│ POST /semantize │
│ GET /check/robots │
└─────────────────────────────┘
Changelog
v1.2.0 (2026-03-29) — Gemini Review Release
- Security: API key authentication (Bearer token via
TRUESOURCE_API_KEY) - Performance: Timeout reduced 180s → 30s (MCP host compatibility)
- Performance: HTML streaming with
</head>early-abort for Schema/OG checks - DX: Improved tool descriptions with explicit trigger words for better LLM routing
- DX:
npx -yinstall flow documented (prevents silent freeze)
v1.1.0 (2026-03-29) — Optimization Release
- Performance: Parallelized 5 checks with
Promise.allSettled(3s → ~1s) - Bugfix: Register
geo_check_robotstool - Bugfix: Fix
{{HOST}}placeholder in generated robots.txt - Bugfix: Fix response format mismatch in check-robots
v1.0.0 (2026-03-28) — Initial Release
- 3 tools:
geo_score,geo_inject,vibetags_generate - Stdio transport, Railway API integration
License
MIT — TrueSource AI / Sascha Deforth
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.