Wolo

Wolo is a minimal Python AI agent CLI tool, similar to Claude Code or opencode. It supports multiple OpenAI-compatible LLM backends and provides a complete toolset for coding tasks.

Features

• Multi-Model Support: Works with any OpenAI-compatible API (OpenAI, Anthropic, DeepSeek, etc.)
• Rich Tool System: Built-in tools for shell commands, file operations, search, and more
• Session Management: Persistent sessions with resume capability
• MCP Integration: Extend capabilities via Model Context Protocol
• Token Compaction: Automatic context management for long sessions
• Path Safety: Configurable write protection for sensitive directories

Installation

One-Click Installation (Recommended)

Linux / macOS:

curl -sSL https://raw.githubusercontent.com/lzjever/wolo/main/scripts/install.sh | bash

Windows (PowerShell):

irm https://raw.githubusercontent.com/lzjever/wolo/main/scripts/install.ps1 | iex

Universal (Python):

python3 -c "$(curl -sSL https://raw.githubusercontent.com/lzjever/wolo/main/scripts/install.py)"

Installation Methods

The installer supports multiple methods:

• auto (default): Automatically selects the best method (uv if available, otherwise pip)
• uv: Uses the uv package manager for faster installation
• pip: Standard pip installation
• source: Installs from git source

# Specify installation method
WOLO_INSTALL_METHOD=uv bash <(curl -sSL https://raw.githubusercontent.com/lzjever/wolo/main/scripts/install.sh)

# Or using the Python installer
python3 <(curl -sSL https://raw.githubusercontent.com/lzjever/wolo/main/scripts/install.py) --method uv

Manual Installation

From PyPI:

# Using uv (recommended)
uv pip install mbos-wolo

# Or using pip
pip install mbos-wolo

From Source:

git clone https://github.com/lzjever/wolo.git
cd wolo

# Using uv
uv sync

# Or using pip
pip install -e .

Quick Start

# Execute a single task
wolo "fix the bug in main.py"

# Pipe context with prompt
git diff | wolo "write commit message"

# Start interactive session
wolo chat

# Resume a session
wolo -r <session-id>

# Wild mode (bypass safety checks/restrictions)
wolo --wild "run unrestricted task"

Configuration

Wolo can be configured via environment variables or config file.

Environment Variables (Recommended for Production):

export WOLO_API_KEY="your-api-key-here"
# or

Config File (Development):

wolo config init

The config file is stored at ~/.wolo/config.yaml:

endpoints:
  - name: default
    model: gpt-4
    api_base: https://api.openai.com/v1
    api_key: your-api-key

# Optional: Enable lexilux client for better compatibility
use_lexilux_client: true

# Optional: MCP server configuration
mcp:
  enabled: true
  servers:
    my-server:
      command: npx
      args: ["-y", "@my/mcp-server"]

Priority: Environment variables take precedence over config file (WOLO_API_KEY > config file).

Commands

Built-in Tools

Development

# Install with dev dependencies
make dev-install

# Run tests
make test

# Run with coverage
make test-cov

# Lint and format
make lint
make format

# Build package
make build

Project Structure

wolo/
├── wolo/              # Main package
│   ├── agent.py       # Agent loop
│   ├── llm_adapter.py # LLM client (lexilux-based)
│   ├── session.py     # Session management
│   ├── tools.py       # Tool implementations
│   ├── cli/           # CLI subsystem
│   ├── compaction/    # Token compaction
│   └── mcp/           # MCP integration
├── tests/             # Test suite
├── docs/              # Documentation
└── pyproject.toml     # Project config

License

Apache-2.0

Contributing

See CONTRIBUTING.md for guidelines.