Fastidious MCP Server
Enables to manage notes and collections in Fastidious AI application through natural language, supporting CRUD operations and search via MCP protocol.
README
Fastidious MCP Server
MCP (Model Context Protocol) server for Fastidious AI notes application.
Deployment
Environment Variables
| Variable | Description | Default |
|---|---|---|
FASTIDIOUS_URL |
Fastidious API base URL | http://localhost:3000 |
Deploy to Coolify
- Create a new service in Coolify
- Connect to this repository
- Set build pack to Nixpacks or Dockerfile
- Configure environment variable:
Or for internal Docker network:FASTIDIOUS_URL=https://blog.tjb.appFASTIDIOUS_URL=http://fastidious:3000 - Set the domain to
mcp.tjb.app - Deploy
Docker
docker build -t fastidious-mcp .
docker run -p 3001:3001 -e FASTIDIOUS_URL=https://blog.tjb.app fastidious-mcp
Usage
Claude Desktop Configuration
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"fastidious": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.tjb.app/sse?token=YOUR_TOKEN_HERE"]
}
}
}
Get your token from Fastidious AI Settings (sidebar → profile → Settings).
Note: Uses mcp-remote to proxy the remote SSE connection.
Local Development (stdio mode)
For local testing with Claude Desktop:
{
"mcpServers": {
"fastidious": {
"command": "npx",
"args": ["tsx", "/path/to/mcp-server/src/index.ts"],
"env": {
"FASTIDIOUS_TOKEN": "YOUR_TOKEN_HERE",
"FASTIDIOUS_URL": "https://blog.tjb.app"
}
}
}
}
Available Tools
| Tool | Description |
|---|---|
create_note |
Create a new markdown note |
get_note |
Get a note by ID |
update_note |
Update a note |
delete_note |
Delete a note |
list_notes |
List all notes (with optional collection filter) |
search_notes |
Search notes by content |
create_collection |
Create a new collection |
get_collection |
Get a collection (with optional contents) |
list_collections |
List all collections |
add_to_collection |
Add notes to a collection |
remove_from_collection |
Remove notes from a collection |
Architecture
Claude Desktop
↓ MCP Protocol (SSE)
MCP HTTP Server (mcp.tjb.app)
↓ HTTP + Bearer Token
Fastidious API (blog.tjb.app/api/mcp/*)
↓
User's Notes & Collections
API Endpoints
GET /health- Health checkGET /sse?token=TOKEN- MCP SSE endpoint for Claude Desktop
Internal Network (Docker/Coolify)
If running alongside Fastidious in the same Docker network:
FASTIDIOUS_URL=http://fastidious:3000
This avoids external network hops for better performance.
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。