Resurgo-MCP
Capture and resurface knowledge - search nodes, quick capture, manage resurfacing schedules
README
Resurgo MCP Server
Official MCP (Model Context Protocol) server for Resurgo - your ideas rise again.
What is Resurgo?
Resurgo helps you capture important information (URLs, notes, ideas) and forces them to resurface at scheduled times - via email digests, browser new tab override, and AI assistant integration. Never lose track of what matters.
Installation
npm install resurgo-mcp
# or
pnpm add resurgo-mcp
Quick Start
Get Your API Key
- Log in to Resurgo
- Go to Settings → API
- Click Create API Key
- Copy your key
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"resurgo": {
"command": "npx",
"args": ["resurgo-mcp"],
"env": {
"RESURGO_API_KEY": "your_api_key_here"
}
}
}
}
Optional: Override the API URL (for self-hosted instances):
{
"mcpServers": {
"resurgo": {
"command": "npx",
"args": ["resurgo-mcp"],
"env": {
"RESURGO_API_KEY": "your_api_key_here",
"RESURGO_API_URL": "https://your-instance.com/api"
}
}
}
}
Available Tools
The MCP server provides 5 tools for managing your knowledge:
1. search_nodes
Search your saved items with full-text search and filters.
{
query: string; // Search term
tags?: string[]; // Filter by tags
status?: 'active' | 'parked' | 'done' | 'archived';
limit?: number; // Max results (default: 20)
}
Example:
"Search my Resurgo nodes for 'react hooks' with the tag 'development'"
2. get_due_items
Get items that are due for review today (your inbox).
{
includeOverdue?: boolean; // Include overdue items (default: true)
}
Example:
"Show me what's due in my Resurgo inbox today"
3. quick_capture
Save a new item with optional scheduling.
{
title: string; // Item title
note?: string; // Optional notes
url?: string; // Optional URL
tags?: string[]; // Tags for organization
resurfaceIn?: string; // When to resurface (see below)
}
Scheduling options:
"tomorrow"- Resurface tomorrow at 9am"next week"or"1 week"- In 7 days"2 weeks"- In 14 days"next month"or"1 month"- In 30 days"2025-06-15"- Specific ISO date
Example:
"Save this URL to Resurgo: https://react.dev/learn with the note 'Review new hooks API' and resurface next week"
4. snooze_node
Defer an item to resurface later.
{
nodeId: string; // Item UUID
until: string; // When to resurface (same format as quick_capture)
}
Example:
"Snooze this Resurgo item until tomorrow"
5. complete_node
Mark an item as done.
{
nodeId: string; // Item UUID
}
Example:
"Mark this Resurgo item as complete"
How It Works
The MCP server connects to your Resurgo account via API and provides AI assistants like Claude with tools to:
- Search through your saved knowledge
- Capture new items during conversations
- Schedule when items should resurface
- Manage your inbox (snooze, complete, etc.)
This means you can have natural conversations like:
"Search my Resurgo notes about TypeScript patterns and create a new note summarizing the key points to review next week"
The AI will use the MCP tools to search, synthesize the information, and create a scheduled note automatically.
Use Cases
- Research Management: Save articles, papers, and links during research sessions
- Learning: Schedule spaced repetition for concepts you're learning
- Project Planning: Capture ideas and schedule them for review at the right time
- Meeting Notes: Save action items and have them resurface before deadlines
- Reading List: Never lose track of articles you want to read
Requirements
- Node.js 20+
- Active Resurgo account (sign up)
- Resurgo API key (create one)
- MCP-compatible AI assistant (Claude Desktop, etc.)
Support
- Issues: GitHub Issues
Privacy & Security
- API Key Authentication: Secure token-based access
- Scoped Access: API keys are tied to your account and can be revoked anytime
- HTTPS Only: All communication encrypted in transit
License
MIT License - see LICENSE for details
Made with ❤️ by Code Atlantic
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。