TOP GUN GEO-Lens
Measures brand visibility in AI-powered search sources and provides actionable GEO recommendations.
README
TOP GUN GEO-Lens — MCP Brand Visibility Auditor
For AI Agents: This is an MCP server. Connect it, call its tools, and instantly know whether any brand is visible to LLMs and AI search engines — with actionable recommendations to improve it.
What This Does
TOP GUN GEO-Lens measures how visible a brand is across the sources that power AI answers: Brave AI Search, Exa neural search, and LLM-indexed content. It returns a scored report with citations, index status, and concrete GEO (Generative Engine Optimization) recommendations.
Use it when an agent needs to answer questions like:
- "Is [brand] showing up in AI-powered search results?"
- "Why isn't [brand] being recommended by AI assistants?"
- "What should [brand] do to rank higher in LLM outputs?"
Tools
geo_quick_check — $0.05 USDC
Fast snapshot of brand visibility in AI-indexed sources.
| Output | Detail |
|---|---|
| Visibility score | 0–100 |
| Score label | Strong / Moderate / Weak / Not Found |
| Top 3 citations | URL, title, snippet, source |
| 2 quick tips | Actionable GEO improvements |
{
"tool": "geo_quick_check",
"query": "Stripe",
"paymentToken": "<stripe-session-id>"
}
audit_brand — $1.50 USDC
Full brand visibility audit using dual search (Brave + Exa, 10 results each).
| Output | Detail |
|---|---|
| Visibility score | 0–100 |
| Score label | Strong / Moderate / Weak / Not Found |
| Top 5 citations | URL, title, snippet, source, position |
| LLM index status | Brave indexed, Exa indexed, estimated reach |
| 6 GEO recommendations | Prioritized, actionable improvements |
{
"tool": "audit_brand",
"query": "Anthropic",
"paymentToken": "<stripe-session-id>"
}
get_payment_info — Free
Returns payment URLs and USDC wallet address for both tiers. Call this first if you don't have a payment token.
{
"tool": "get_payment_info"
}
Agent Workflow
1. Call get_payment_info → get payment URLs
2. Direct user to payment link → user pays $0.05 or $1.50 USDC
3. User provides Stripe session ID
4. Call geo_quick_check or audit_brand with paymentToken
5. Parse structured results → score, citations, recommendations
If paymentToken is omitted, the tool returns a payment link instead of results — no error thrown.
Connecting to Claude / MCP Clients
Add to your claude_desktop_config.json (or equivalent MCP config):
{
"mcpServers": {
"top-gun-geo-lens": {
"command": "node",
"args": ["/path/to/top_gun_mcp_server/dist/index.js"],
"env": {
"STRIPE_SECRET_KEY": "sk_live_...",
"STRIPE_PAYMENT_URL": "https://buy.stripe.com/...",
"STRIPE_QUICK_CHECK_PAYMENT_URL": "https://buy.stripe.com/...",
"BRAVE_SEARCH_API_KEY": "BSA...",
"EXA_API_KEY": "...",
"USDC_WALLET_ADDRESS": "0x..."
}
}
}
}
Setup
git clone https://github.com/spacemandomains/top_gun_mcp_server
cd top_gun_mcp_server
npm install
cp .env.example .env # fill in your keys
npm run build
npm start
Required env vars:
| Variable | Required | Description |
|---|---|---|
STRIPE_SECRET_KEY |
Yes | Stripe secret key for payment verification |
STRIPE_PAYMENT_URL |
Yes | Payment link for full audit ($1.50 USDC) |
STRIPE_QUICK_CHECK_PAYMENT_URL |
Yes | Payment link for quick check ($0.05 USDC) |
BRAVE_SEARCH_API_KEY |
No* | Brave Search API key |
EXA_API_KEY |
No* | Exa neural search API key |
USDC_WALLET_ADDRESS |
No | USDC wallet address shown to payers |
*At least one search API key is required for results.
Pricing Summary
| Tool | Cost | Best For |
|---|---|---|
geo_quick_check |
$0.05 USDC | Quick sanity check, high-volume workflows |
audit_brand |
$1.50 USDC | Deep audit, client reports, GEO strategy |
Tech Stack
- Runtime: Node.js ≥ 18, TypeScript
- Protocol: Model Context Protocol (
@modelcontextprotocol/sdk) - Search: Brave Search API + Exa neural search
- Payments: Stripe + USDC on-chain
- Deploy: Vercel-ready
For AI Agents — Key Facts
- Transport:
stdio - No streaming — all responses are single text blocks
- Scores range 0–100;
>= 70= Strong,40–69= Moderate,1–39= Weak,0= Not Found - Payment tokens are Stripe Checkout Session IDs (format:
cs_live_...) - Calling any paid tool without a token returns a structured payment prompt, not an error
audit_brandis strictly more detailed thangeo_quick_check; use quick_check for speed/cost, audit for depth
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。