ds_task_marketing_assistant_ai/docs/API_Documendation.md

# API Documentation

## Base URL
```
http://localhost:8000
```

## Endpoints

### Generate Marketing Copy
**POST** `/generate`

Generate marketing copy based on prompt and brand guidelines.

#### Request Body
```json
{
  "prompt": "string",
  "content_type": "general",
  "tone": "string"
}
```

#### Response
```json
"Generated marketing copy text"
```

### Add Campaign
**POST** `/add-campaign`

Add a marketing campaign to the vector store for future reference.

#### Request Body
```json
{
  "content": "string",
  "content_type": "string",
  "metadata": {}
}
```

### Search Campaigns
**GET** `/search`

Search for similar campaigns using vector similarity.

#### Query Parameters
- `query` (string): Search query
- `content_type` (string, optional): Filter by content type
- `k` (int, optional): Number of results to return (default: 3)

#### Response
```json
[
  {
    "content": "string",
    "content_type": "string",
    "metadata": {},
    "_search_distance": 0.0
  }
]
```

### Health Check
**GET** `/health`

Check API health status.

#### Response
```json
{
  "status": "healthy"
}
```

### Root
**GET** `/`

API root endpoint with basic information.

## Content Types
- `general` - General marketing content
- `email` - Email campaigns
- `social` - Social media posts
- `ad` - Advertisement copy
- `blog` - Blog content

## Error Responses
All endpoints may return:
- **422 Validation Error** - Invalid request format
- **500 Internal Server Error** - Server error

```json
{
  "detail": [
    {
      "loc": ["field_name"],
      "msg": "Error message",
      "type": "error_type"
    }
  ]
}
```

## Vector Store Features
- Uses `all-MiniLM-L6-v2` model for embeddings (384 dimensions)
- FAISS index for fast similarity search
- Persistent storage of campaigns
- Automatic index rebuilding on startup