strands-clova
CLOVA Studio model provider for Strands Agents SDK
Features
- Native Korean Language Support - Optimized for Korean context and nuances
- Bilingual Capabilities - Seamlessly handle both Korean and English
- Full Streaming Support - Real-time response streaming with SSE
- Simple Authentication - Just needs CLOVA_API_KEY
- Easy Integration - Drop-in replacement for any Strands model provider
- Type Safe - Full type hints and mypy support
Requirements
- Python 3.10+
- Strands Agents SDK 1.7.0+
- CLOVA Studio API key from Naver Cloud Platform
Installation
pip install strands-agents strands-clova
Quick Start
Basic Usage
from strands_clova import ClovaModel
from strands import Agent
# Initialize CLOVA model
model = ClovaModel(
api_key="your-clova-api-key", # or set CLOVA_API_KEY env var
model="HCX-005",
temperature=0.7,
max_tokens=2048
)
# Create an agent
agent = Agent(model=model)
# Use the agent
response = await agent.invoke_async("안녕하세요! 오늘 날씨가 어떤가요?")
print(response.message)
Streaming Responses
import asyncio
from strands_clova import ClovaModel
async def stream_example():
model = ClovaModel(api_key="your-api-key")
async for event in model.stream("한국의 전통 음식 3가지를 소개해주세요"):
if event.get("type") == "text":
print(event["text"], end="", flush=True)
asyncio.run(stream_example())
With System Prompt
model = ClovaModel(api_key="your-api-key")
async for event in model.stream(
"Python 리스트 컴프리헨션을 설명해주세요",
system_prompt="You are a helpful coding assistant. Provide concise answers."
):
# Process streaming events
pass
Configuration
Environment Variables
export CLOVA_API_KEY="your-api-key"
export CLOVA_REQUEST_ID="optional-request-id" # For request tracking
Model Parameters
model = ClovaModel(
model="HCX-005", # Model ID (currently only HCX-005 is supported)
temperature=0.7, # Sampling temperature (0.0-1.0)
max_tokens=4096, # Maximum tokens to generate
top_p=0.8, # Nucleus sampling parameter
top_k=0, # Top-k sampling parameter
repeat_penalty=1.1, # Repetition penalty
stop=["\\n\\n"], # Stop sequences
)
Dynamic Configuration
# Update configuration at runtime
model.update_config(
temperature=0.2,
max_tokens=1024
)
# Get current configuration
config = model.get_config()
print(config)
Examples
Korean Customer Support Bot
from strands_clova import ClovaModel
from strands import Agent
model = ClovaModel(api_key="your-api-key", temperature=0.3)
agent = Agent(
model=model,
system_prompt="당신은 친절한 고객 서비스 상담원입니다."
)
response = await agent.invoke_async(
"제품 반품 절차를 알려주세요"
)
Bilingual Document Processing
# Process Korean document and get English summary
response = await agent.invoke_async(
"다음 한국어 문서를 영어로 요약해주세요: [문서 내용]"
)
Creative Content Generation
model = ClovaModel(temperature=0.9, max_tokens=2048)
agent = Agent(model=model)
response = await agent.invoke_async(
"AI가 교육에 미치는 영향에 대한 블로그 포스트를 작성해주세요"
)
Testing
Run Unit Tests
pip install -e ".[dev]"
pytest tests/unit/ -v
Run Integration Tests
Requires CLOVA_API_KEY environment variable:
export CLOVA_API_KEY="your-api-key"
pytest tests/integration/ -v
API Reference
ClovaModel
Main model class that implements the Strands Model abstract base class.
Methods
stream(messages, tool_specs, system_prompt, **kwargs)- Stream responsesget_config()- Get current configurationupdate_config(**kwargs)- Update configurationstructured_output()- Not yet implemented (raises NotImplementedError)
ClovaModelError
Exception raised for CLOVA-specific errors.
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add some amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
License
This project is licensed under the MIT License - see the LICENSE file for details.
Links
Acknowledgments
- Thanks to the Strands Agents team for creating an excellent SDK
- Thanks to Naver for providing CLOVA Studio API
Status
- Basic streaming support - Complete
- Korean and English language support - Complete
- System prompts - Complete
- Configuration management - Complete
- Structured output support - In Progress
- Tool/Function calling - In Progress
Made for the Korean AI community
Metadata
Release files for strands-clova 0.1.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| strands_clova-0.1.2.tar.gz | 12.4 kB | Without provenance | Details for strands_clova-0.1.2.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| strands_clova-0.1.2-py3-none-any.whl | Python 3 | none | any | Details for strands_clova-0.1.2-py3-none-any.whl |
Total release size: 21.1 kB
Release files / strands_clova-0.1.2.tar.gz
| Download URL | strands_clova-0.1.2.tar.gz |
|---|---|
| Size | 12.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
0a1f5c9fc1051888793de162b8c258e3e2a67939422e3fcacc4627587072fb19
|
|
BLAKE2b-256 checksum How to use checksums |
fa374b4463f529963d6204272b5247f302f4b9a851d67e713ccd3622fedc9c69
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.7
|
Release files / strands_clova-0.1.2-py3-none-any.whl
| Download URL | strands_clova-0.1.2-py3-none-any.whl |
|---|---|
| Size | 8.7 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
7e934b84340d701b5beec499752605b6fb86e62e125f6afe21e99f20c6d08b0b
|
|
BLAKE2b-256 checksum How to use checksums |
92ae80bf7683103bfd1136ffe23afc864f404cbc676b5ceeadcacfb05b5cdcca
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.7
|
Security for strands-clova 0.1.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.