# Lifecycle MCP Usage Examples
## Understanding the Setup
The lifecycle-mcp server can be used in different ways depending on your needs:
### 1. **One Server, Multiple Projects** (Recommended)
Install once, use everywhere. Each project gets its own database.
```bash
# Install lifecycle-mcp globally (do this once)
cd /Users/you/repos/lifecycle-mcp
pip install -e .
# Project A - E-commerce site
cd /Users/you/projects/ecommerce-site
claude mcp add lifecycle lifecycle-mcp -e LIFECYCLE_DB=./lifecycle.db
# Creates: /Users/you/projects/ecommerce-site/lifecycle.db
# Project B - Mobile app
cd /Users/you/projects/mobile-app
claude mcp add lifecycle lifecycle-mcp -e LIFECYCLE_DB=./lifecycle.db
# Creates: /Users/you/projects/mobile-app/lifecycle.db
# Each project has its own separate lifecycle database!
```
### 2. **Shared Database Across Projects**
Use one database for related projects:
```bash
# Create a shared database location
mkdir -p ~/lifecycle-data
# Add server with shared database
claude mcp add lifecycle lifecycle-mcp -e LIFECYCLE_DB=~/lifecycle-data/shared.db
# All projects using this configuration share the same database
```
### 3. **Project-Specific Virtual Environment**
For complete isolation:
```bash
# Create virtual environment in your project
cd /Users/you/projects/my-app
python -m venv .venv
source .venv/bin/activate
# Install lifecycle-mcp in this venv
pip install -e /path/to/lifecycle-mcp
# Add using the venv's Python
claude mcp add lifecycle .venv/bin/lifecycle-mcp -e LIFECYCLE_DB=./lifecycle.db
```
## Common Scenarios
### Scenario 1: Team Development
Your team clones the lifecycle-mcp repo to a standard location:
```bash
# Team standard location
TEAM_MCP_PATH="/opt/mcp-servers/lifecycle-mcp"
# Each developer adds it to their Claude
claude mcp add lifecycle $(which python) $TEAM_MCP_PATH/server.py -e LIFECYCLE_DB=./lifecycle.db
```
### Scenario 2: Personal Projects
Install globally for all your personal projects:
```bash
# One-time setup
pip install --user -e ~/repos/lifecycle-mcp
# Use in any project
cd ~/projects/anything
claude mcp add lifecycle lifecycle-mcp -e LIFECYCLE_DB=./lifecycle.db
```
### Scenario 3: Testing Different Versions
Run directly without installation:
```bash
# Clone specific version/branch
git clone -b feature-xyz https://github.com/heffrey78/lifecycle-mcp.git lifecycle-test
cd lifecycle-test
# Run directly with uv
claude mcp add lifecycle-test $(which uv) -- --directory $(pwd) run server.py -e LIFECYCLE_DB=./test.db
```
## Troubleshooting
### "Command not found: lifecycle-mcp"
The command isn't in your PATH. Solutions:
1. Find where it's installed:
```bash
pip show -f lifecycle-mcp | grep lifecycle-mcp
```
2. Use the full path:
```bash
claude mcp add lifecycle /full/path/to/lifecycle-mcp -e LIFECYCLE_DB=./lifecycle.db
```
3. Or run from source:
```bash
claude mcp add lifecycle $(which python) /path/to/lifecycle-mcp/server.py -e LIFECYCLE_DB=./lifecycle.db
```
### Multiple Python Versions
Be specific about which Python to use:
```bash
# Use specific Python version
claude mcp add lifecycle python3.11 /path/to/lifecycle-mcp/server.py -e LIFECYCLE_DB=./lifecycle.db
# Or with pyenv
claude mcp add lifecycle $(pyenv which python) /path/to/lifecycle-mcp/server.py -e LIFECYCLE_DB=./lifecycle.db
```
### Database Location Best Practices
1. **Per-project database** (default):
```bash
LIFECYCLE_DB=./lifecycle.db # Creates in current directory
```
2. **Centralized databases**:
```bash
LIFECYCLE_DB=~/Documents/lifecycle-dbs/project-name.db
```
3. **Shared team database**:
```bash
LIFECYCLE_DB=/shared/team/lifecycle.db
```
## Configuration File Examples
### Basic Configuration
`~/.config/claude/claude_desktop_config.json`:
```json
{
"mcpServers": {
"lifecycle": {
"command": "lifecycle-mcp",
"env": {
"LIFECYCLE_DB": "./lifecycle.db"
}
}
}
}
```
### Advanced Configuration
```json
{
"mcpServers": {
"lifecycle-dev": {
"command": "/Users/me/repos/lifecycle-mcp/.venv/bin/lifecycle-mcp",
"env": {
"LIFECYCLE_DB": "./lifecycle-dev.db",
"LOG_LEVEL": "DEBUG"
}
},
"lifecycle-prod": {
"command": "lifecycle-mcp",
"env": {
"LIFECYCLE_DB": "/Users/me/Documents/projects/production.db"
}
}
}
}
```
This allows you to have multiple configurations for different purposes!
