# MCP Configuration Templates
This document contains the configuration templates needed to connect the MIST (Model Intelligence System for Tasks) MCP server to different AI assistants and code editors.
## Environment Configuration
Before setting up your AI assistant connections, make sure you have created a `.env` file in the MIST root directory:
```env
# Notes storage configuration
MIST_NOTES_DIR=C:/Users/username/Documents/Notes
# Google API configuration
MIST_GOOGLE_CREDENTIALS_PATH=./credentials.json
MIST_GOOGLE_TOKEN_PATH=./token.json
# Optional debugging configuration
MIST_LOG_LEVEL=INFO
```
## 1. Claude Desktop Configuration
Add the following configuration to your Claude Desktop settings to connect to the MIST MCP server:
### Windows Example
```json
{
"mcpServers": {
"M.I.S.T.": {
"command": "uv",
"args": [
"run",
"--with",
"mcp[cli]",
"--with-editable",
"C:/Users/username/projects/mist",
"mcp",
"run",
"C:/Users/username/projects/mist/server.py"
],
"env": {
"MIST_GOOGLE_CREDENTIALS_PATH": "C:/Users/username/projects/mist/credentials.json",
"MIST_GOOGLE_TOKEN_PATH": "C:/Users/username/projects/mist/token.json",
"MIST_NOTES_DIR": "C:/Users/username/Documents/Notes"
}
}
}
}
```
### macOS/Linux Example
```json
{
"mcpServers": {
"M.I.S.T.": {
"command": "uv",
"args": [
"run",
"--with",
"mcp[cli]",
"--with-editable",
"/home/username/projects/mist",
"mcp",
"run",
"/home/username/projects/mist/server.py"
],
"env": {
"MIST_GOOGLE_CREDENTIALS_PATH": "/home/username/projects/mist/credentials.json",
"MIST_GOOGLE_TOKEN_PATH": "/home/username/projects/mist/token.json",
"MIST_NOTES_DIR": "/home/username/Documents/Notes"
}
}
}
}
```
## 2. Zed Editor Configuration
Add the following configuration to your Zed editor settings to connect to the MIST MCP server:
### Windows Example
```json
"context_servers": {
"mist-server": {
"command": {
"path": "uv",
"args": [
"run",
"--with",
"mcp[cli]",
"--with-editable",
"C:/Users/username/projects/mist",
"mcp",
"run",
"C:/Users/username/projects/mist/server.py"
],
"env": {
"MIST_GOOGLE_CREDENTIALS_PATH": "C:/Users/username/projects/mist/credentials.json",
"MIST_GOOGLE_TOKEN_PATH": "C:/Users/username/projects/mist/token.json",
"MIST_NOTES_DIR": "C:/Users/username/Documents/Notes"
}
},
"settings": {
"enable_server": true
}
}
}
```
### macOS/Linux Example
```json
"context_servers": {
"mist-server": {
"command": {
"path": "uv",
"args": [
"run",
"--with",
"mcp[cli]",
"--with-editable",
"/home/username/projects/mist",
"mcp",
"run",
"/home/username/projects/mist/server.py"
],
"env": {
"MIST_GOOGLE_CREDENTIALS_PATH": "/home/username/projects/mist/credentials.json",
"MIST_GOOGLE_TOKEN_PATH": "/home/username/projects/mist/token.json",
"MIST_NOTES_DIR": "/home/username/Documents/Notes"
}
},
"settings": {
"enable_server": true
}
}
}
```
## 3. Running with Direct Command
You can also run MIST directly from the command line:
### Windows (Command Prompt)
```cmd
set MIST_NOTES_DIR=C:\Users\username\Documents\Notes
set MIST_GOOGLE_CREDENTIALS_PATH=C:\Users\username\projects\mist\credentials.json
set MIST_GOOGLE_TOKEN_PATH=C:\Users\username\projects\mist\token.json
cd C:\Users\username\projects\mist
uv run --with mcp[cli] --with-editable . mcp run mist/server.py
```
### Windows (PowerShell)
```powershell
$env:MIST_NOTES_DIR="C:\Users\username\Documents\Notes"
$env:MIST_GOOGLE_CREDENTIALS_PATH="C:\Users\username\projects\mist\credentials.json"
$env:MIST_GOOGLE_TOKEN_PATH="C:\Users\username\projects\mist\token.json"
cd C:\Users\username\projects\mist
uv run --with mcp[cli] --with-editable . mcp run mist/server.py
```
### macOS/Linux (Bash/Zsh)
```bash
export MIST_NOTES_DIR=/home/username/Documents/Notes
export MIST_GOOGLE_CREDENTIALS_PATH=/home/username/projects/mist/credentials.json
export MIST_GOOGLE_TOKEN_PATH=/home/username/projects/mist/token.json
cd /home/username/projects/mist
uv run --with mcp[cli] --with-editable . mcp run mist/server.py
```
