Aden Aggregates MCP Server
Exposes Aden Aggregates submittal data as tools callable by Claude, enabling retrieval of location, material, pricing, and lab report information to build submittal packages.
README
Aden Aggregates MCP Server
Model Context Protocol server that exposes Aden Aggregates submittal data as tools callable by Claude.
Environment Variables (Railway)
In your Railway project → Variables, add:
| Variable | Value |
|---|---|
SUPABASE_URL |
https://hmlspqsjuyybjkkexdwd.supabase.co |
SUPABASE_SERVICE_ROLE_KEY |
eyJhbGci... (your service role key) |
Railway automatically injects PORT.
Deploy to Railway
- Push this repo to GitHub
- Go to railway.app → New Project → Deploy from GitHub repo
- Select this repo — Railway auto-detects Node.js and runs
npm start - After deploy, go to Settings → Networking → Generate Domain to get your public URL
Your server URL will look like: https://aden-mcp-production.up.railway.app
Register in Claude.ai
- Go to claude.ai → Settings → Integrations
- Click Add MCP Server
- Paste your Railway URL +
/sse:https://aden-mcp-production.up.railway.app/sse - Name it
Aden Aggregatesand save
Available Tools
| Tool | Description |
|---|---|
list_locations |
All quarry/yard codes and names |
list_materials |
All materials with specs |
list_sales_reps |
Active sales reps |
get_location |
Full location record by code |
get_material |
Material record by name |
get_price |
2026 FOB price for location + material |
get_lab_reports |
Exhibit records for location + material |
get_cover_letter_template |
Cover letter template text |
get_sales_rep |
Rep record by partial name |
lookup_customer |
Search customers by name or code |
Updated Skill System Prompt
Replace the hardcoded data in your Claude.ai skill with this condensed version:
You are the Aden Aggregates Submittal Package Assistant. You have access to MCP tools that provide all location, material, pricing, lab report, and customer data — always call the tools rather than guessing.
WORKFLOW:
1. Ask: location, material, customer, sales rep, project name, spec reference
2. Call get_location, get_material, get_price, get_lab_reports, get_cover_letter_template with the provided inputs
3. Build the cover letter: fill {material}, {location_name}, {location_phone}, {location_address}, {specification} into the template_text returned by get_cover_letter_template
4. List exhibits in order (EXHIBIT A, EXHIBIT B…) exactly as returned by get_lab_reports
5. Output the complete package as formatted text ready for PDF generation
RULES:
- Never invent prices, specs, or certifications — always fetch from tools
- If get_price returns indicative=true, note "Price indicative — confirm with sales"
- Cover letter must be on Aden Aggregates letterhead format (company name, address, date, RE: line, body, signature block)
- Signature block: [Sales Rep Name] | Aden Aggregates | [rep email] | [rep phone]
- NYSDOT status from get_location goes in the certification line verbatim
- PDF generation: after user confirms content, output a /generate command with all fields as JSON
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。