What environment variables are required?
`TRILIUM_API_TOKEN` is required. `TRILIUM_API_URL` defaults to `http://localhost:8080/etapi`. `PERMISSIONS` (default `READ;WRITE`) and `VERBOSE` (default `false`) are optional.
@tan-yong-sheng
基本信息
MCP server for TriliumNext Notes
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"triliumnext-mcp": {
"command": "npx",
"args": [
"triliumnext-mcp"
],
"env": {
"TRILIUM_API_URL": "http://localhost:8080/etapi",
"TRILIUM_API_TOKEN": "<YOUR_TRILIUM_API_TOKEN>",
"PERMISSIONS": "READ;WRITE"
}
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
点击上方按钮,在临时沙箱中运行此服务器并实时获取工具列表。
A Model Context Protocol (MCP) server for TriliumNext Notes. It provides tools to search, create, update, delete, and manage notes and attributes in your Trilium Notes instance via the ETAPI. This prototype is intended for developer use; backup your notes before using.
Install via npx triliumnext-mcp or run with Docker (pre‑built image from GitHub Container Registry). Configure environment variables: TRILIUM_API_URL (default http://localhost:8080/etapi), TRILIUM_API_TOKEN (required), PERMISSIONS (optional, default READ;WRITE), and VERBOSE (optional). Add the server configuration to your MCP client (e.g., Claude Desktop, Claude Code, Cursor, Cline) using stdio transport.
`TRILIUM_API_TOKEN` is required. `TRILIUM_API_URL` defaults to `http://localhost:8080/etapi`. `PERMISSIONS` (default `READ;WRITE`) and `VERBOSE` (default `false`) are optional.
No – it is a prototype for developer use only. Always back up your Trilium notes before using this tool.
It interacts with your own Trilium Notes instance via the ETAPI. All data remains in your Trilium instance; no external storage is used.
The server uses stdio transport and follows the standard MCP protocol, making it compatible with any MCP client.

Model Context Protocol Servers
High-performance code intelligence MCP server. Indexes codebases into a persistent knowledge graph — average repo in milliseconds. 162 languages, sub-ms queries, 99% fewer tokens. Single static binary, zero dependencies.

Shared memory for AI coding agents. Claude Code, Cursor and Codex read your project's decisions, lessons and tasks before they act.

One permissioned folder your team and your AI agents both log into.

Local-first MCP server that snapshots your project's working state into one context object. Open-core, privacy-first.
评论