sven-client 0.2.0.dev0

Sven CLI client for agent automation

Sven

Sven is the CLI client for ActiveAgent, providing a command-line interface for interacting with the ActiveAgent server.

Installation

pip install sven

Usage

# Basic usage
sven --help

# Connect to a server
sven client --url http://localhost:8000

# Run an agent
sven agent run --name my-agent

# List available tools
sven tools list

# Store question-answer pairs
sven qa store --file examples/qa_example.json

# Query the question-answer database
sven qa query "How do I create a branch in Git?"

Configuration

Sven can be configured using a YAML configuration file. Create a .sven.yml file in your home directory or project directory:

api_url: "https://api.swedishembedded.com"
api_port: 443
api_key: "your-api-key-here"
model: "claude-3-7-sonnet-latest"
persona: "coder"

See the configuration documentation for more details on how to configure Sven and provide API keys.

Features

Question-Answer Database

Sven provides commands to interact with a semantic question-answer database:

# Store question-answer pairs from a JSON or YAML file
sven qa store --file examples/qa_example.json

# Query the database using semantic search
sven qa query "How do I create a branch in Git?"

The questions and answers are organized into categories and can include metadata. See the example files in the examples/ directory for the correct format:

• JSON Example
• YAML Example

Development

Setup

# Clone the repository
git clone https://git.swedishembedded.com/core/platform/sven.git
cd sven

# Create a virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install development dependencies
pip install -e ".[dev]"

# Install pre-commit hooks
pre-commit install

Testing

pytest

Linting

# Run all linters
pre-commit run --all-files

# Run individual linters
black .
isort .
mypy .
ruff .

CI/CD Pipeline

This project uses GitLab CI/CD for automated testing, building, and deployment. The pipeline includes:

• Lint: Code quality checks with black, isort, ruff, and mypy
• Test: Unit and integration tests with pytest
• Build: Package building with setuptools
• Publish: Publishing to PyPI (on tags) and TestPyPI (on main branch)

Releasing

We use semantic versioning for this project. To create a new release:

1. Update the CHANGELOG.md with your changes
2. Use bump2version to increment the version:

   # For a patch release (0.1.0 -> 0.1.1)
   bump2version patch
   
   # For a minor release (0.1.0 -> 0.2.0)
   bump2version minor
   
   # For a major release (0.1.0 -> 1.0.0)
   bump2version major
   

3. Push the new tag to trigger the release pipeline:

   git push --tags
   

Contributing

Please see CONTRIBUTING.md for details on how to contribute to this project.

License

MIT