familiar-add-tool
Scaffold and register new sensor, actuator, or service tools for familiar-ai, automating file creation and boilerplate integration in agent.py and config.py.
Introduction
The familiar-add-tool utility is a specialized developer tool designed to streamline the expansion of the familiar-ai agent's capabilities. Whether you are integrating a new physical sensor like a microphone or GPS, adding a robotic actuator such as an arm or LED controller, or wrapping an external third-party API service, this skill provides a standardized scaffolding workflow. By automating the repetitive aspects of tool development, it ensures consistency across the agent's modular architecture.
- Automatically generates tool class files in src/familiar_agent/tools/ using a robust template.
- Provides built-in methods for tool definition (Anthropic input_schema format) and execution routing.
- Seamlessly updates agent.py to handle imports, initialization, and tool definitions.
- Configures necessary environment variables in config.py and .env.example files.
- Automates the integration of new tools into the agent's system prompt to ensure the LLM correctly understands how to invoke the new functionality.
This utility is intended for developers and power users building custom extensions for their familiar-ai companion. When using this skill, ensure that the tool name follows snake_case naming conventions. The process requires identifying the tool's core verbs (actions the agent can perform), required environment configuration variables, and a clear description for the system prompt. After execution, it is recommended to run linters such as ruff to ensure the generated code adheres to project standards. This skill significantly reduces the friction of expanding the agent's physical or digital interface, allowing for faster iteration in robotics and home automation integration.
Repository Stats
- Stars
- 81
- Forks
- 7
- Open Issues
- 2
- Language
- Python
- Default Branch
- main
- Sync Status
- Idle
- Last Synced
- May 3, 2026, 05:19 PM