kiarina-lib-anthropic 1.18.2

Anthropic API client library for kiarina namespace

kiarina-lib-anthropic

A Python library for Anthropic API integration with configuration management using pydantic-settings-manager.

Features

• Configuration Management: Use pydantic-settings-manager for flexible configuration
• Type Safety: Full type hints and Pydantic validation
• Secure Credential Handling: API keys are protected using SecretStr
• Multiple Configurations: Support for multiple named configurations (e.g., different projects/environments)
• Environment Variable Support: Configure via environment variables with KIARINA_LIB_ANTHROPIC_ prefix
• Custom Base URL: Support for custom Anthropic-compatible API endpoints

Installation

pip install kiarina-lib-anthropic

Quick Start

Basic Usage

from kiarina.lib.anthropic import AnthropicSettings, settings_manager

# Configure Anthropic API
settings_manager.user_config = {
    "default": {
        "api_key": "sk-ant-your-api-key-here"
    }
}

# Get settings
settings = settings_manager.settings
print(f"API Key configured: {settings.api_key.get_secret_value()[:10]}...")

Environment Variable Configuration

Configure authentication using environment variables:

export KIARINA_LIB_ANTHROPIC_API_KEY="sk-ant-your-api-key-here"

from kiarina.lib.anthropic import settings_manager

# Settings are automatically loaded from environment variables
settings = settings_manager.settings
print(f"API Key configured: {settings.api_key.get_secret_value()[:10]}...")

Multiple Configurations

Manage multiple Anthropic configurations (e.g., different projects or environments):

from kiarina.lib.anthropic import settings_manager

# Configure multiple projects
settings_manager.user_config = {
    "project_a": {
        "api_key": "sk-ant-project-a-key"
    },
    "project_b": {
        "api_key": "sk-ant-project-b-key"
    }
}

# Switch between configurations
settings_manager.active_key = "project_a"
project_a_settings = settings_manager.settings
print(f"Project A API Key: {project_a_settings.api_key.get_secret_value()[:10]}...")

settings_manager.active_key = "project_b"
project_b_settings = settings_manager.settings
print(f"Project B API Key: {project_b_settings.api_key.get_secret_value()[:10]}...")

Custom Base URL

Use with Anthropic-compatible APIs:

from kiarina.lib.anthropic import settings_manager

settings_manager.user_config = {
    "custom": {
        "api_key": "your-custom-key",
        "base_url": "https://custom.anthropic-compatible.api.com"
    }
}

settings = settings_manager.settings
print(f"Base URL: {settings.base_url}")

Configuration

This library uses pydantic-settings-manager for flexible configuration management.

AnthropicSettings

The AnthropicSettings class provides the following configuration fields:

Field	Type	Required	Default	Description
api_key	SecretStr	Yes	-	Anthropic API key (masked in logs)
base_url	str | None	No	None	Custom base URL for Anthropic-compatible APIs

Environment Variables

All settings can be configured via environment variables with the KIARINA_LIB_ANTHROPIC_ prefix:

# API Key (required)
export KIARINA_LIB_ANTHROPIC_API_KEY="sk-ant-your-api-key"

# Custom Base URL (optional)
export KIARINA_LIB_ANTHROPIC_BASE_URL="https://api.anthropic.com"

Programmatic Configuration

from pydantic import SecretStr
from kiarina.lib.anthropic import AnthropicSettings, settings_manager

# Direct settings object
settings = AnthropicSettings(
    api_key=SecretStr("sk-ant-your-api-key")
)

# Via settings manager
settings_manager.user_config = {
    "default": {
        "api_key": "sk-ant-your-api-key"  # Automatically converted to SecretStr
    }
}

Runtime Overrides

from kiarina.lib.anthropic import settings_manager

# Override specific settings at runtime
settings_manager.cli_args = {
    "base_url": "https://custom.api.com"
}

settings = settings_manager.settings
print(f"Base URL: {settings.base_url}")  # Uses overridden value

Security

API Key Protection

API keys are stored using Pydantic's SecretStr type, which provides the following security benefits:

• Masked in logs: Keys are displayed as ********** in string representations
• Prevents accidental exposure: Keys won't appear in debug output or error messages
• Explicit access required: Must use .get_secret_value() to access the actual key

from kiarina.lib.anthropic import settings_manager

settings = settings_manager.settings

# API key is masked in string representation
print(settings)  # api_key=SecretStr('**********')

# Explicit access to get the actual key
api_key = settings.api_key.get_secret_value()

API Reference

AnthropicSettings

class AnthropicSettings(BaseSettings):
    api_key: SecretStr
    base_url: str | None = None

Pydantic settings model for Anthropic API configuration.

Fields:

• api_key (SecretStr): Anthropic API key (protected)
• base_url (str | None): Optional custom base URL for Anthropic-compatible APIs

settings_manager

settings_manager: SettingsManager[AnthropicSettings]

Global settings manager instance for Anthropic configuration. See: pydantic-settings-manager

Development

Prerequisites

• Python 3.12+

Setup

# Clone the repository
git clone https://github.com/kiarina/kiarina-python.git
cd kiarina-python

# Setup development environment
mise run setup

Running Tests

# Run format, lint, type checks and tests
mise run package kiarina-lib-anthropic

# Coverage report
mise run package:test kiarina-lib-anthropic --coverage

Dependencies

• pydantic-settings - Settings management
• pydantic-settings-manager - Advanced settings management

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributing

This is a personal project, but contributions are welcome! Please feel free to submit issues or pull requests.

Related Projects

• kiarina-python - The main monorepo containing this package
• pydantic-settings-manager - Configuration management library used by this package
• kiarina-lib-openai - Similar library for OpenAI API