# Installation
## Requirements
- Python 3.11+
- [uv](https://docs.astral.sh/uv/) package manager
- Ollama (optional, for CLI LLM features)
## Install Delia
```bash
# Clone the repository
git clone https://github.com/zbrdc/delia.git
cd delia
# Install dependencies
uv sync
```
## Configure MCP Clients
Auto-detect and configure all supported AI clients:
```bash
uv run delia install
```
Or install to a specific client:
```bash
uv run delia install claude # Claude Code
uv run delia install cursor # Cursor
uv run delia install vscode # VS Code
```
List available clients:
```bash
uv run delia install --list
```
## Start the Server
### Option A: HTTP Transport (Recommended)
Best for multi-project setups. One server handles all projects.
```bash
uv run delia run -t http --port 8765
```
Add to each project's `.mcp.json`:
```json
{
"mcpServers": {
"delia": {
"type": "http",
"url": "http://localhost:8765/mcp"
}
}
}
```
> **Note**: HTTP servers won't appear in Claude Code's `/mcp` list, but tools work normally.
### Option B: stdio Transport
Per-project server, managed by the AI client. Shows in `/mcp` list.
```json
{
"mcpServers": {
"delia": {
"command": "uv",
"args": ["--directory", "/path/to/delia", "run", "delia", "serve"]
}
}
}
```
## Initialize Your Project
### Via MCP (Recommended)
Let the AI agent initialize the project:
```
# In Claude Code or Cursor, ask:
"Initialize this project with Delia"
# Or use the MCP tool directly:
project(action="init", path="/path/to/your-project")
```
### Via CLI (requires Ollama)
If you have Ollama running locally:
```bash
cd ~/your-project
uv run --directory /path/to/delia delia init-project
```
This creates `.delia/` with playbooks tailored to your tech stack.
## Verify Setup
```bash
uv run delia doctor
```
## Next Steps
- [Quick Start](quick-start.md) - 5-minute guide
- [Configuration](configuration.md) - Customize settings
- [Workflow](../user-guide/workflow.md) - Learn the learning loop
