AI 어시스턴트가 Mind Elixir Desktop 애플리케이션 내에서 마인드맵을 직접 생성, 편집 및 정리할 수 있습니다.
다른 이름 mind-elixir/mcp, mcp, @mind-elixir/mcp
- 스타
- 1
- Tools
- 11
- 다운로드, 30일
- —
- 최신 버전
- 0.1.5
- 마지막 커밋
- 2026년 9월 15일
- 라이선스
- MIT
Mind Elixir MCP란?
Mind Elixir은(는) MCP server입니다. AI 어시스턴트가 Mind Elixir Desktop 애플리케이션 내에서 마인드맵을 직접 생성, 편집 및 정리할 수 있습니다. 11개 도구를 제공합니다. 예: mindmap_data, node_id, topic.
프로젝트 README에서 생성된 요약
Mind Elixir MCP server 정보
Mind Elixir MCP Server는 AI 어시스턴트와 Mind Elixir Desktop 애플리케이션 사이를 연결하는 브리지 역할을 합니다. 도구 호출을 로컬 HTTP 엔드포인트로 전달함으로써 Claude Desktop, Cursor, VS Code, Windsurf와 같은 stdio 전용 client를 데스크톱 앱에 연결할 수 있습니다. 이 설정을 통해 AI가 데스크톱 소프트웨어에서 현재 작업 중인 마인드맵과 직접 상호작용할 수 있습니다.
사용자는 7가지 특정 도구를 통해 시각적 구조를 관리할 수 있습니다. 이러한 도구를 사용하면 어시스턴트가 완전히 새로운 맵을 생성하고, 기존 토픽의 이름을 변경하며, 자식 노드를 추가하고, 심지어 멀리 떨어진 노드 사이에 라벨이 붙은 화살표를 그려 관계를 표시할 수도 있습니다. 또한 이 server는 형제 노드 그룹을 하나의 아이디어로 축소하는 요약 노드 추가와 같은 고급 정리 기능도 지원합니다.
이를 사용하려면 기기에 Mind Elixir Desktop이 설치되어 실행 중이어야 합니다. 브리지는 기본적으로 6595 포트를 통해 앱에 연결됩니다. 브리지가 stdio에서 HTTP로의 변환을 처리하지만, 직접 streamable HTTP 연결이 가능한 client는 브리지를 거치지 않고 데스크톱 앱에 직접 연결할 수 있습니다.
활용 사례
- 1문서를 맵으로 변환AI가 긴 텍스트 문서나 대화 내용을 구조화된 시각적 마인드맵으로 변환할 수 있습니다.
- 2프로젝트 구조 정리개발자가 어시스턴트에게 복잡한 프로젝트를 공식적인 작업 분할 구조(WBS)로 나누도록 지시할 수 있습니다.
- 3회의 주제 요약어시스턴트를 사용하여 회의 녹취록에서 주요 주제를 추출하고 맵으로 구성할 수 있습니다.
- 4기존 노트 정교화AI가 이전에 수동으로 작성한 노트를 재구성하고 정리할 수 있습니다.
프로젝트의 README를 바탕으로 작성되었습니다. 모든 활용 사례는 해당 내용의 인용문과 대조하여 확인합니다.
Mind Elixir MCP 설치 방법
형식 확인: 2026년 11월 10일Cursor에 추가Goose에 추가Claude Code에 Mind Elixir MCP 설치
$ claude mcp add mcp -- npx -y @mind-elixir/mcpCursor에 Mind Elixir MCP 설치
~/.cursor/mcp.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@mind-elixir/mcp"
]
}
}
}VS Code에 Mind Elixir MCP 설치
.vscode/mcp.json
{
"servers": {
"mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@mind-elixir/mcp"
]
}
}
}VS Code CLI에 Mind Elixir MCP 설치
$ code --add-mcp '{"name":"mcp","type":"stdio","command":"npx","args":["-y","@mind-elixir/mcp"]}'Claude Desktop에 Mind Elixir MCP 설치
claude_desktop_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@mind-elixir/mcp"
]
}
}
}Codex에 Mind Elixir MCP 설치
$ codex mcp add mcp -- npx -y @mind-elixir/mcpCodex config.toml에 Mind Elixir MCP 설치
~/.codex/config.toml
[mcp_servers.mcp] command = "npx" args = ["-y", "@mind-elixir/mcp"]
Gemini CLI에 Mind Elixir MCP 설치
$ gemini mcp add -s user mcp npx -- -y @mind-elixir/mcpGemini settings.json에 Mind Elixir MCP 설치
~/.gemini/settings.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@mind-elixir/mcp"
]
}
}
}Mind Elixir MCP 도구 11
README에서 감지됨mindmap_datanode_idtopicparent_idtextparentstartendlabelfrombidirectional
| mindmap_data | — | |
| node_id | — | |
| topic | — | |
| parent_id | — | |
| text | — | |
| parent | — | |
| start | — | |
| end | — | |
| label | — | |
| from | — | |
| bidirectional | — |
구성
MIND_ELIXIR_MCP_URLTarget URL of the Mind Elixir Desktop MCP endpoint. Defaults to http://127.0.0.1:6595/mcp
| 이름 | 설명 | 유형 |
|---|---|---|
| MIND_ELIXIR_MCP_URL | Target URL of the Mind Elixir Desktop MCP endpoint. Defaults to http://127.0.0.1:6595/mcp |
자동 검사
2026년 11월 14일에 실행됨전체 보고서 →자동 검사를 통과했습니다. 이는 보안 감사가 아닙니다. 코드를 직접 읽고 서버에 필요한 최소한의 권한만 부여하십시오. 검사 방식
- 소스 저장소 연결 및 접근 가능mind-elixir/mcp is public and reachablegithub.com/mind-elixir/mcp ↗통과
- 패키지가 이 저장소를 소스로 명시@mind-elixir/mcp names mind-elixir/mcp as sourcenpmjs.com/package/@mind-elixir/mcp ↗통과
- 빌드 출처(Sigstore 증명)No build attestation published for this release검사되지 않음
- 오픈 소스 라이선스 (OSI)MITspdx.org/licenses/MIT.html ↗통과
- 최근 1년 내 커밋 또는 릴리스Last commit or release 2026-09-15github.com/mind-elixir/mcp ↗통과
- OpenSSF Scorecard검사되지 않음
- 패키지 이름이 인기 패키지와 유사하지 않음No popular package one edit away from @mind-elixir/mcp통과
- 패키지 및 해당 종속성에 중요한 보안 권고 없음99 resolved dependencies checked; no critical advisoryosv.dev ↗통과
- 모델에 대한 숨겨진 지침 없음11 tool descriptions and the README read; no override phrases, credential paths or hidden characters통과
- 파괴적인 도구에 라벨 표시됨No tool named like delete, drop or exec통과
- 원격 엔드포인트가 MCP 핸드셰이크에 응답검사되지 않음
README
GitHub에서 보기Mind Elixir MCP Server
Let AI assistants build and edit real mind maps in Mind Elixir Desktop.
This is the stdio ↔ HTTP bridge that connects stdio-only MCP clients — Claude Desktop, Claude Code, Cursor, VS Code, Cline, Roo Code, Windsurf — to the MCP server built into the Mind Elixir Desktop app.
Requires Mind Elixir Desktop to be installed and running. This server is a local controller for the desktop app, not a standalone cloud service. The bridge forwards tool calls to
http://127.0.0.1:6595/mcpand streams the results back.
- Registry name:
io.github.mind-elixir/mcp - npm package:
@mind-elixir/mcp - Setup guide: https://app.mind-elixir.com/blog/use-mcp
Contents
- What you can do with it
- Requirements
- Quick start
- Client configuration
- Connecting without the bridge (HTTP)
- Tools reference
- Recommended agent workflow
- CLI reference
- How it works
- Troubleshooting
- Security and privacy
- Development
- License
What you can do with it
Seven tools let an assistant read and mutate the mind map that is currently open in the app:
- Start a fresh map, or replace the current one with a structure the model designed.
- Inspect every topic, node ID, and parent/child relationship in the current map.
- Rename nodes, add child nodes, and get back the ID of what was just created.
- Attach summary nodes that collapse a range of siblings into one idea.
- Draw labelled arrows between any two nodes to express relationships that the tree cannot.
Typical uses: turning a conversation or a document into a structured map, breaking a project down into a work breakdown structure, summarising a meeting into themes, or having an assistant reorganise notes you already made by hand.
Requirements
| Requirement | Detail |
|---|---|
| Mind Elixir Desktop | Installed and running. The app hosts the actual MCP server on 127.0.0.1:6595. |
| Node.js | 18 or newer — only needed to run the bridge via npx, npm i -g, or a local install. |
| MCP client | Any client that speaks MCP over stdio, or one that can connect directly to a Streamable HTTP endpoint. |
| Port | 6595 must be free on loopback. Override with --port if it is not. |
Quick start
Start Mind Elixir Desktop first, then verify the bridge can reach it:
npx -y @mind-elixir/mcp@latest
A running bridge prints nothing to stdout (that channel is reserved for JSON-RPC) and stays in the foreground. Diagnostics go to stderr. Quit it with Ctrl+C.
To install the CLI globally instead:
npm install -g @mind-elixir/mcp
mind-elixir-mcp
Then configure your MCP client — see the next section.
Client configuration
Claude Desktop
Edit claude_desktop_config.json:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"mind-elixir": {
"command": "npx",
"args": ["-y", "@mind-elixir/mcp@latest"]
}
}
}
Restart Claude Desktop after saving.
Claude Code
claude mcp add mind-elixir -- npx -y @mind-elixir/mcp@latest
Cursor
Create or edit ~/.cursor/mcp.json (or use Settings → Features → MCP → + Add New MCP Server):
{
"mcpServers": {
"mind-elixir": {
"command": "npx",
"args": ["-y", "@mind-elixir/mcp@latest"]
}
}
}
VS Code (GitHub Copilot agent mode)
Add to .vscode/mcp.json in your workspace, or to your user-level mcp.json:
{
"servers": {
"mind-elixir": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@mind-elixir/mcp@latest"]
}
}
}
Cline, Roo Code, Windsurf
These extensions share the mcpServers shape used by Cursor:
{
"mcpServers": {
"mind-elixir": {
"command": "npx",
"args": ["-y", "@mind-elixir/mcp@latest"]
}
}
}
Non-default port
If Mind Elixir Desktop is listening somewhere other than 6595, pass the port through:
{
"mcpServers": {
"mind-elixir": {
"command": "npx",
"args": ["-y", "@mind-elixir/mcp@latest", "--port", "7000"]
}
}
}
Connecting without the bridge (HTTP)
If your client can talk Streamable HTTP directly, skip this package entirely and point it at the app:
{
"mcpServers": {
"mind-elixir": {
"type": "http",
"url": "http://127.0.0.1:6595/mcp"
}
}
}
The app exposes:
| Endpoint | Method | Purpose |
|---|---|---|
/mcp |
POST / GET |
MCP Streamable HTTP endpoint (the one you want). |
/ping |
GET |
Liveness probe; returns pong. |
CORS is enabled for the Mind Elixir Cloud origin, so browser-hosted clients can call the local server too.
Tools reference
All tools return a plain-text result. Successful mutations answer with OK (or, for add_child, the new node's ID); failures come back as a string beginning with Error:.
| Tool | Purpose | Returns |
|---|---|---|
new_mindmap |
Create a blank map and switch the app to the editor | OK |
get_all_nodes |
Read the whole current map | Node tree JSON |
generate_mindmap |
Replace the current map from structured data | OK |
edit_topic |
Rename a node | OK |
add_child |
Add a child node | Done. New Node's ID is <id>. |
add_node_summary |
Summarise a range of siblings | OK |
add_arrow |
Connect two nodes with a labelled arrow | OK |
Every tool has a 15-second operation budget. If the app does not report back in time, the tool returns Timeout waiting for mindmap operation to complete.
new_mindmap
Creates a new, empty mind map and navigates the app to its editor page.
No parameters.
get_all_nodes
Returns the current map as a nested node tree — every topic, its ID, and its children.
No parameters. Shape of the response:
{
"topic": "Root Topic",
"id": "1",
"children": [
{ "topic": "Child Topic", "id": "b3f1c0e2-…", "children": [] }
]
}
Always call this before editing or adding to a map you did not create in this session — IDs are the only way to address nodes.
generate_mindmap
Replaces the entire current map with the structure you supply.
| Parameter | Type | Description |
|---|---|---|
mindmap_data |
string | A JSON string with a nodeData root object. |
{
"nodeData": {
"topic": "Root Topic",
"id": "1",
"children": [
{ "topic": "Child Topic 1", "id": "1-1", "children": [] },
{ "topic": "Child Topic 2", "id": "1-2" }
]
}
}
Requirements:
- Every node needs
topic(string) andid(string). childrenis optional.- Use hierarchical numbering (
1,1-1,1-2,2,2-1, …) so later tools can address nodes predictably.
edit_topic
Changes a node's text in place, preserving its position, children, summaries, and arrows.
| Parameter | Type | Description |
|---|---|---|
node_id |
string | ID of the node to rename, e.g. 1, 1-1, or a UUID. |
topic |
string | The new topic text. |
Returns Error: Node not found if node_id does not exist — call get_all_nodes to refresh your view.
add_child
Adds a child node under an existing parent. The app generates the ID, so the result carries it.
| Parameter | Type | Description |
|---|---|---|
parent_id |
string | ID of the parent node. |
topic |
string | Topic text for the new child. |
Returns Done. New Node's ID is <id>. — keep that ID if you plan to edit the node later. Returns Error: Parent node not found when the parent is missing.
add_node_summary
Attaches a summary node that consolidates a contiguous range of a parent's children.
| Parameter | Type | Description |
|---|---|---|
text |
string | The summary text. |
parent |
string | ID of the parent whose children are being summarised. |
start |
integer | Zero-based index of the first child to include. |
end |
integer | Zero-based index of the last child to include (inclusive). |
Example — summarise the first three children of node 1:
{ "text": "These three all concern onboarding", "parent": "1", "start": 0, "end": 2 }
add_arrow
Draws a labelled connection between two nodes, for relationships the tree cannot express.
| Parameter | Type | Description |
|---|---|---|
label |
string | Text shown on or near the arrow, e.g. leads to, depends on. |
from |
string | ID of the source node. |
to |
string | ID of the target node. |
bidirectional |
boolean | true for ↔, false for →. |
Recommended agent workflow
- Orient first. Call
get_all_nodesto learn the real IDs and structure before touching anything. Never assume an ID. - Decide between build and patch. For a brand-new map use
generate_mindmapwith hierarchical IDs. For incremental work on an existing map useadd_childandedit_topic. - Capture generated IDs.
add_childreturns a UUID — record it if the node will be edited again. - Do not rely on the app being on the editor page. Any tool call brings the app to the front and opens the editor; if it was not already editing a map, a new blank map is created first. So always confirm the current state with
get_all_nodesbefore a write. - Keep calls sequential. Mutations are applied in order, and there is no undo exposed over MCP.
CLI reference
USAGE:
npx -y @mind-elixir/mcp [options]
OPTIONS:
-u, --url <url> Target Mind Elixir MCP URL (default: http://127.0.0.1:6595/mcp)
-p, --port <port> Change target port (default: 6595)
-t, --transport <sse|http> Force transport mode ('sse' or 'http', default: auto)
-v, --version Show version
-h, --help Show this help message
ENVIRONMENT VARIABLES:
MIND_ELIXIR_MCP_URL Target URL to connect to
The package installs two equivalent binaries: mind-elixir-mcp and mcp.
Transport selection in auto mode: if the target URL path ends with /mcp, Streamable HTTP is used; otherwise the bridge falls back to the legacy SSE transport. Force one with --transport.
How it works
+--------------------------+
| Claude Desktop / Cursor |
+------------+-------------+
| stdio (stdin / stdout), JSON-RPC
v
+------------+-------------+
| @mind-elixir/mcp | bridge process
+------------+-------------+
| Streamable HTTP -> http://127.0.0.1:6595/mcp
v
+------------+-------------+
| Mind Elixir Desktop | Tauri app, owns the tools
+--------------------------+
- The bridge translates stdio JSON-RPC messages into HTTP requests and streams responses back verbatim.
- Startup is tolerant. If the app is still booting, messages are buffered and replayed in strict FIFO order once the connection is established. The bridge retries the connection three times, one second apart.
- Protocol version is negotiated. The bridge starts at
2024-11-05and adopts whatever version theinitializehandshake settles on, forwarding it to the app. - Failures are explicit. If the app never becomes reachable, every buffered request receives a JSON-RPC error (
-32000) readingMind Elixir Desktop is not running. Please open Mind Elixir Desktop and try again. (<url>),stderrgets a human-readable explanation, and the process exits with code1. Errors raised during an operation are returned to the caller as their original message rather than being swallowed.
Troubleshooting
| Symptom | Cause and fix |
|---|---|
Mind Elixir Desktop is not running |
Launch the desktop app, then reconnect the MCP server in your client. |
Tool call returns Timeout waiting for mindmap operation to complete |
The app did not respond within 15 s — usually a modal dialog or a busy window. Bring the app to the front and retry. |
Connection refused on 6595 |
The port is taken by another process, or the app is on a different port. Restart the app, or pass --port <port> / set MIND_ELIXIR_MCP_URL. |
| The client lists no tools | Confirm the bridge starts by running npx -y @mind-elixir/mcp@latest in a terminal, then restart the client so it re-reads its MCP config. |
| A blank map appeared unexpectedly | The app was not on an editor page when a tool ran, so it created one. Read the map with get_all_nodes before writing. |
Error: Node not found / Error: Parent node not found |
The ID is stale or comes from a different map. Re-read with get_all_nodes. |
| Nothing appears in the client's log | All bridge diagnostics go to stderr; stdout carries protocol traffic only. |
Security and privacy
- The app's MCP server binds
127.0.0.1only — it is never exposed to your network, and it has no authentication. Any local process on your machine can drive the app through it, so treat port6595like any other local control surface. - The bridge is a local child process of your MCP client. It reads and writes nothing on disk; the only network traffic is the loopback HTTP call to the app.
- Everything happens on your machine. No mind map content, prompt, or telemetry is sent to a server by this package or by the desktop app's MCP endpoint.
- Grant the server the same care you would give any tool that edits your documents: it can rewrite the map you have open. Prefer clients configured to ask before running tools when you are working on something important.
Development
Requires Node.js 18+ and pnpm 11+ (the allowBuilds entry in pnpm-workspace.yaml depends on it).
pnpm install
pnpm build # tsup -> dist/index.js
pnpm dev # tsup --watch
The build emits a single bundled ESM file with a #!/usr/bin/env node shebang; prepublishOnly rebuilds before every npm publish.
License
MIT © Mind Elixir
Mind Elixir MCP 질문
Mind Elixir MCP를 Claude Code에 추가하려면 어떻게 하나요?
터미널에서 claude mcp add mcp -- npx -y @mind-elixir/mcp를 실행합니다. 모든 프로젝트에서 사용할 수 있게 하려면 --scope user를 추가합니다.
Mind Elixir MCP를 Cursor에 추가하려면 어떻게 하나요?
~/.cursor/mcp.json를 열고 mcpServers 아래에 서버를 추가합니다. 위 설치 섹션에 표시된 Cursor 구성을 사용합니다.
Mind Elixir MCP는 어떤 도구를 제공하나요?
11개 도구를 제공합니다. 예: mindmap_data, node_id, topic, parent_id, text.
이 server를 사용하기 위한 요구 사항은 무엇인가요?
Mind Elixir Desktop이 설치 및 실행 중이어야 하며, Node.js 버전 18 이상, 그리고 stdio 또는 streamable HTTP를 지원하는 MCP client가 필요합니다.
AI 어시스턴트가 마인드맵으로 실제로 무엇을 할 수 있나요?
어시스턴트는 새 맵을 시작하고, 모든 노드와 관계를 검사하며, 노드 이름을 변경하고, 자식 노드를 추가하고, 요약 노드를 첨부하며, 노드 사이에 라벨이 붙은 화살표를 그릴 수 있습니다.