ftc-knowledge-server
Remote MCP server for FTC & Pedro Pathing research, providing tools to search technical notes, code templates, and verified team examples to assist AI agents in code generation and robotics development.
README
ftc-knowledge-server
Remote MCP (Model Context Protocol) server for FTC + Pedro Pathing research. Deploy on Vercel and connect from Cursor, Claude Code, Claude Desktop, or any MCP-compatible AI agent.
What's included
- Technical gold notes (Pinpoint, OTOS, Pedro gotchas)
- AI agent code generation guidance
- Pedro Pathing patterns (pathing, localization, autos, teleop, tuning)
- Java code templates (FConstants, LConstants, Auto, Subsystem)
- Verified team repository index
- High-signal Java file index
- Official docs and community links
Deploy to Vercel
Or manually:
npm install
npm run build
npx vercel --prod
Connect to Cursor
Add to .cursor/mcp.json in your project (or Cursor Settings → MCP):
{
"mcpServers": {
"ftc-knowledge": {
"url": "https://YOUR-DEPLOYMENT.vercel.app/api/mcp"
}
}
}
If your Cursor version doesn't support remote URL MCP yet:
{
"mcpServers": {
"ftc-knowledge": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://YOUR-DEPLOYMENT.vercel.app/api/mcp"]
}
}
}
Restart Cursor after saving. The AI can now call tools like ftc_get_full_context and ftc_search_research.
Connect to Claude Code / Claude Desktop
Add to ~/.claude/claude_desktop_config.json or project MCP config:
{
"mcpServers": {
"ftc-knowledge": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://YOUR-DEPLOYMENT.vercel.app/api/mcp"]
}
}
}
MCP Tools
| Tool | Description |
|---|---|
ftc_search_research |
Search notes, gotchas, docs, verified resources |
ftc_get_template |
Get FConstants, LConstants, Auto, Subsystem templates |
ftc_get_full_context |
Full context bundle for code generation |
ftc_get_patterns |
Patterns by category (pathing, localization, etc.) |
ftc_get_team_examples |
Verified team repo index |
ftc_search_codebase |
Search high-signal Java files index |
REST API
For apps that can't use MCP directly:
GET /api/context?action=info
GET /api/context?action=search&q=pinpoint
GET /api/context?action=template&type=fconstants
GET /api/context?action=context
GET /api/context?action=patterns&category=pathing
GET /api/context?action=teams
GET /api/context?action=codebase&pattern=Follower
Example prompts (after connecting)
Use ftc_get_full_context to load FTC context, then write an autonomous OpMode with Pedro Pathing.
Search ftc_search_research for the PID zeroing bug and fix my FConstants.
Use ftc_get_template with type fconstants for a Pinpoint mecanum setup.
Local development
npm install
npm run dev
# MCP endpoint: http://localhost:3000/api/mcp
# Landing page: http://localhost:3000
License
Research corpus compiled from public FTC/Pedro Pathing sources. Use for educational and team programming purposes.
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。