OORB CLI – Robotics coding Assistant

OORB (Open Organic Robotics Benchmark) is an open-source initiative focused on advancing robotics through AI-powered tools and benchmarks.

OORB CLI is a command-line assistant designed to answer questions and assist with tasks related to ROS2 (Robot Operating System 2). Powered by modern language models and enhanced with vector search and retrieval-augmented generation, it serves as an intelligent tool for roboticists and developers working with ROS2.

---

Features

• Focused exclusively on ROS2
• Interactive and one-shot Q&A modes
• Backend support for OpenAI, Azure OpenAI, and Ollama
• Local inference support with Ollama
• Retrieval-augmented answers with Milvus vector store
• Configurable ROS distro support - Specify your target ROS2 distribution
• Custom retrieval endpoint support - Use your own vector database or API
• Configurable model, temperature, and backend preferences
• Tool calling capabilities with access to:
  • read_file - Read and analyze files
  • write_file - Create and modify files
  • search_in_files - Search across file contents
  • run_command - Execute system commands
  • list_directory - Browse directory structures
  • edit_file_lines - Make precise file edits

Recent Enhancements

• ROS Distro Selection: Set your preferred ROS2 distribution (humble, jazzy, rolling, etc.)
• Custom Retrieval Endpoints: Connect to your own vector database or knowledge API
• Enhanced Configuration: Multiple ways to configure settings via environment variables or CLI options
• Endpoint Testing: Built-in tools to test and validate retrieval API connectivity

Note:

You can find a quick demo of the CLI on the following link : link

---

Installation

Option 1: Install from PyPI (Recommended)

pip install oorb-cli

The CLI is available on PyPI and can be installed directly via pip.

Option 2: Development Installation

1. Clone the repository

git clone https://github.com/OORB-Open-Organic-Robotics/oorb-cli
cd oorb-cli

2. Create virtual environment and install dependencies using uv

uv venv
source .venv/bin/activate
uv pip install -e .

This installs the CLI in editable mode, allowing you to make changes to the source code without reinstalling.

Note: uv pip install -e . is functionally equivalent to pip install -e ., but uses uv for faster dependency resolution and isolated builds.

---

Configuration

Set the appropriate API key depending on the backend you intend to use.

OpenAI

export OPENAI_API_KEY="your-key"

Azure OpenAI

export AZURE_OPENAI_API_KEY="your-key"
export AZURE_OPENAI_ENDPOINT="your-endpoint"

ROS Distro Configuration

Specify your target ROS2 distribution for enhanced retrieval and distro-specific answers:

# Set default ROS distro (overrides auto-detection)
export OORB_ROS_DISTRO=humble

# Or use the standard ROS environment variable
export ROS_DISTRO=humble

Supported ROS2 Distributions:

• humble (LTS) - Ubuntu 20.04, 22.04
• iron - Ubuntu 22.04
• jazzy (LTS) - Ubuntu 24.04
• rolling - Latest development version
• Legacy ROS1: noetic, melodic

Custom Retrieval Endpoint

Configure your own vector database or retrieval API:

# Use custom retrieval endpoint
export OORB_RETRIEVAL_ENDPOINT="http://your-server:8000"

# Or use the standard configuration
export RETRIEVAL_API_BASE_URL="http://your-server:8000"

---

Usage

Once installed, the CLI becomes available via the oorb command (or python oorb_cli.py if unlinked).

Main Commands

chat - Interactive ROS2 Assistant

Start an interactive chat session or ask a single question with full customization options.

Basic Usage:

# Interactive mode
oorb chat

# Single question mode
oorb chat -p "How do I create a custom message type in ROS2?"

# With specific ROS distro
oorb chat --distro humble -p "create a publisher node"
oorb chat -d jazzy

# With custom retrieval endpoint
oorb chat --retrieval-endpoint http://localhost:9000 -p "setup navigation stack"

Available Options:

• -b, --backend - Choose LLM backend: azure, openai, or ollama
• -m, --model - Specify model name (e.g., gpt-4o-mini, gpt-3.5-turbo)
• -t, --temperature - Set response creativity (0.0-1.0, default: 0.3)
• -d, --distro - ROS distro to use (e.g., humble, jazzy, rolling)
• -r, --retrieval-endpoint - Custom retrieval API endpoint URL
• -p, --prompt - Single prompt for non-interactive mode
• --use-tools/--no-tools - Enable/disable tool calling (default: auto-detect)

Examples:

# Use specific backend and model with ROS distro
oorb chat -b openai -m gpt-4o-mini -d humble -t 0.7

# Quick question with custom settings
oorb chat -b azure -m gpt-3.5-turbo -d jazzy -p "Create a ROS2 launch file for navigation"

# Use custom retrieval endpoint
oorb chat -r http://my-vector-db:8000 -d rolling -p "Setup SLAM toolbox"

# Disable tools for faster responses
oorb chat --no-tools -d humble -p "What is a ROS2 node?"

list-distros - View Supported ROS Distributions

Display all supported ROS distributions with detailed information.

Basic Usage:

oorb list-distros

What it shows:

• All supported ROS/ROS2 distributions
• LTS (Long Term Support) status
• Compatible Ubuntu versions
• Current vs. available distros
• Configuration instructions

test-retrieval - Test Retrieval Endpoint

Test connectivity and features of retrieval API endpoints.

Basic Usage:

# Test current configured endpoint
oorb test-retrieval

# Test specific endpoint
oorb test-retrieval --endpoint http://localhost:9000

Available Options:

• -e, --endpoint - Specific endpoint URL to test

What it shows:

• Endpoint accessibility status
• Available API endpoints (e.g., /retrieve, /health, /docs)
• Configuration suggestions
• Connection diagnostics

list-models - View Available Models

Display all available models for different backends.

Basic Usage:

oorb list-models

Available Options:

• -b, --backend - Filter by specific backend: azure, openai, or ollama

Examples:

# List all models
oorb list-models

# List only OpenAI models
oorb list-models -b openai

# List only local Ollama models
oorb list-models -b ollama

status - System Health Check

Check system configuration, API keys, and service availability.

Basic Usage:

oorb status

What it shows:

• API key configuration status
• Ollama service status and available models
• Retrieval API connectivity and endpoint information
• Current ROS distro detection
• Available backends
• Configuration warnings and errors
• Quick setup recommendations

---

Advanced Configuration

Priority Order for Settings

OORB CLI uses the following priority order for configuration:

ROS Distro Selection:

1. --distro command line parameter
2. OORB_ROS_DISTRO environment variable
3. ROS_DISTRO environment variable (standard ROS)
4. Auto-detection from installed ROS distributions

Retrieval Endpoint Selection:

1. --retrieval-endpoint command line parameter
2. OORB_RETRIEVAL_ENDPOINT environment variable
3. RETRIEVAL_API_BASE_URL environment variable
4. Default: http://localhost:8000

Configuration Examples

# Set persistent defaults via environment
export OORB_ROS_DISTRO=humble
export OORB_RETRIEVAL_ENDPOINT=http://my-vector-db:8000
export OPENAI_API_KEY=your-api-key

# Override defaults for specific commands
oorb chat -d jazzy -r http://localhost:9000 -p "your question"

# Test your configuration
oorb status
oorb test-retrieval
oorb list-distros

---

Contributing

We welcome contributions to the OORB CLI project! Please see our CONTRIBUTING.md file for detailed guidelines on how to contribute to the project.

---

License

This project is open-source and available under the appropriate license. Please check the LICENSE file for details.