Skip to main content

A Python SDK for PromptStudio

Project description

PromptStudio Python SDK

A Python SDK for interacting with PromptStudio API and AI platforms directly.

Installation

From PyPI

pip install promptstudio-sdk

From Source

git clone https://github.com/your-repo/promptstudio-sdk.git
cd promptstudio-sdk
pip install -e .

Development Setup

  1. Create a virtual environment:
python -m venv venv
  1. Activate the virtual environment:
# On Windows
venv\Scripts\activate

# On Unix or MacOS
source venv/bin/activate
  1. Install dependencies:
pip install -r requirements.txt

Usage

Initializing the SDK

from promptstudio_sdk import PromptStudio

client = PromptStudio({
    'api_key': 'YOUR_API_KEY',
    'env': 'prod',  # Use 'prod' for production environment
    'bypass': True ,
    'is_logging': True
})

Configuration Options

bypass (default: False)

The bypass parameter determines whether to use the local AI provider directly or route requests through PromptStudio's API:

  • When bypass=True: Requests go directly to the AI provider, bypassing PromptStudio's API
  • When bypass=False: Requests are routed through PromptStudio's API for additional processing and logging

is_session_enabled (default: True)

The is_session_enabled parameter controls whether the chat maintains conversation history:

  • When True: The conversation history is maintained between requests using the provided session_id
  • When False: Each request is treated independently with no conversation history

is_logging (default: True)

The is_logging parameter controls whether to log chat interactions:

  • When True: Chat interactions are logged for analysis and tracking
  • When False: Chat interactions are not logged

shot (default: -1)

The shot parameter controls how many message pairs to include from the beginning of the conversation:

  • When -1: All previous messages are included (default behavior)
  • When 0: No previous messages are included
  • When n > 0: Include the first n pairs of messages (2n messages total)

For example:

  • With shot=2: Includes the first 4 messages (2 pairs from the start)
  • With shot=3: Includes the first 6 messages (3 pairs from the start)
# Example: Using shot to include first 2 pairs of messages
response = client.chat_with_prompt(
    prompt_id="your_prompt_id",
    user_message=[{"type": "text", "text": "Hello"}],
    session_id="test_session",
    shot=2  # Include first 2 pairs of messages (4 messages total from the start)
)

# If your conversation history has 10 messages:
# shot=2 will use messages[0:4] (first 4 messages)
# shot=0 will use no previous messages
# shot=-1 will use all 10 messages

Chatting with a Prompt

response = client.chat_with_prompt(
    prompt_id="your_prompt_id",
    user_message=[
        {
            "type": "text",
            "text": "Hello, how are you?"
        }
    ],
    memory_type="fullMemory",
    window_size=0,
    session_id="test_session",
    variables={},
    is_session_enabled=True,
    shot=2,
)

print(response)

Complete Example

from promptstudio_sdk import PromptStudio

def main():
    # Initialize the client
    client = PromptStudio({
        'api_key': 'YOUR_API_KEY',
        'env': 'test',
        'bypass': True,
        'is_logging': True
    })

    try:
        # Get all prompts
        prompts = client.get_all_prompts("your_folder_id")
        print("Available prompts:", prompts)

        # Chat with a specific prompt
        response = client.chat_with_prompt(
            prompt_id="your_prompt_id",
            user_message=[
                {
                    "type": "text",
                    "text": "Hello, how are you?"
                }
            ],
            memory_type="windowMemory",
            window_size=10,
            session_id="test_session",
            variables={},
            is_session_enabled=True,
            shot=2,
        )
        print("Chat response:", response)

    except Exception as e:
        print(f"An error occurred: {e}")

if __name__ == "__main__":
    main()

Testing

Setting Up Tests

  1. Install test dependencies:
pip install pytest pytest-cov
  1. Create a .env file in the root directory with your test credentials:
PROMPTSTUDIO_API_KEY=your_test_api_key
PROMPTSTUDIO_ENV=test  # Use 'prod' for production environment

Running Tests

Run all tests:

pytest

Run tests with coverage:

pytest --cov=promptstudio_sdk

Writing Tests

Create test files in the tests directory. Here's an example test:

import pytest
from promptstudio_sdk import PromptStudio

def test_chat_with_prompt():
    client = PromptStudio({
        'api_key': 'test_api_key',
        'env': 'test',
        'bypass': True,
        'is_logging': True
    })

    response = client.chat_with_prompt(
        prompt_id: str,
        user_message: List[Dict[str, Union[str, Dict[str, str]]]],
        memory_type: str,
        window_size: int,
        session_id: str,
        variables: Dict[str, str],
        version: Optional[int] = None,
        is_session_enabled: Optional[bool] = True,
        shot: Optional[int] = -1,
    )

    assert isinstance(response, dict)
    assert 'response' in response

Type Hints

The SDK uses Python type hints for better IDE support and code documentation. Here are some key types:

from typing import Dict, List, Union, Optional

# Message types
ImageMessage = Dict[str, Union[str, Dict[str, str]]]  # {"type": "image_url", "image_url": {"url": "..."}}
TextMessage = Dict[str, str]  # {"type": "text", "text": "..."}
UserMessage = List[Union[ImageMessage, TextMessage]]

# Memory types
Memory = Literal["fullMemory", "windowMemory", "summarizedMemory"]

These are the only valid memory types that can be used with the `memory_type` parameter:
- `"fullMemory"`: Maintains the complete conversation history
- `"windowMemory"`: Keeps a sliding window of recent messages
- `"summarizedMemory"`: Maintains a summarized version of the conversation history

# Request payload
RequestPayload = Dict[str, Union[UserMessage, Memory, int, str, Dict[str, str], Optional[int]]]

Error Handling

The SDK raises exceptions for various error cases:

from promptstudio_sdk import PromptStudio

try:
    client = PromptStudio({
        'api_key': 'YOUR_API_KEY_from_promptstudio',
        'env': 'test',
        'bypass': True
    })
    response = client.chat_with_prompt(...)
except requests.exceptions.HTTPError as e:
    print(f"HTTP error occurred: {e}")
except requests.exceptions.RequestException as e:
    print(f"Network error occurred: {e}")
except Exception as e:
    print(f"An error occurred: {e}")

Contributing

  1. Fork the repository
  2. Create a new branch for your feature
  3. Make your changes
  4. Run the tests to ensure everything works
  5. Submit a pull request

License

This SDK is released under the MIT License.

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

promptstudio_sdk-1.0.247.tar.gz (39.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

promptstudio_sdk-1.0.247-py3-none-any.whl (44.9 kB view details)

Uploaded Python 3

File details

Details for the file promptstudio_sdk-1.0.247.tar.gz.

File metadata

  • Download URL: promptstudio_sdk-1.0.247.tar.gz
  • Upload date:
  • Size: 39.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.2

File hashes

Hashes for promptstudio_sdk-1.0.247.tar.gz
Algorithm Hash digest
SHA256 5dba9981c31a5fa0690a0aa4b21408bac95b68ab0c4ca91b8d361047ab759091
MD5 6c4214ab2486ac50523a5013870bd2da
BLAKE2b-256 ff79e06b584b4c36e13b742c7b4531abcb34aad6f4a8876833a9e51443d5a027

See more details on using hashes here.

File details

Details for the file promptstudio_sdk-1.0.247-py3-none-any.whl.

File metadata

File hashes

Hashes for promptstudio_sdk-1.0.247-py3-none-any.whl
Algorithm Hash digest
SHA256 40ee1d36b8fafeeee136b5481ddecb11ca6bfc001edee550169aaaabbea832f8
MD5 1ce1b32fc1fed5c618255e6c9dcfa196
BLAKE2b-256 5cd5a46121cb337832e76b9e4a0aeff45d1974357f03c8773d7b5f4bbf348e95

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page