RasaHQ/rasa-sdk

SDK for the development of custom actions for Rasa

What it solves

This project provides a Python SDK that allows developers to create custom actions for Rasa, enabling chatbots to perform complex tasks beyond simple predefined responses.

How it works

Developers use the rasa-sdk package to implement custom logic in Python. These actions are then hosted on an action server (which can be run via Docker) that communicates with the Rasa core via a webhook at http://localhost:5055/webhook.

Who it’s for

It is designed for developers building conversational AI assistants using the Rasa framework.

Highlights

  • Python-based custom action development
  • Official Docker image available for easy deployment
  • Support for custom system and PyPI dependencies via Dockerfile extensions
  • Integrated linting and formatting using Ruff

Related

  • Project
    OpenHands/software-agent-sdk

    A Python and REST API SDK for building AI agents that can write, edit, and maintain code in local or ephemeral workspaces.

  • Project
    robocorp/robocorp

    A Python-based automation stack that enables the creation of automation tasks and AI Actions to interact with web browsers, Windows apps, and data files.

  • Project
    boston-dynamics/spot-sdk

    A software development kit for the Spot robot that allows developers to create custom applications and sensor payloads to extend the robot's capabilities.

  • Project
    RasaHQ/rasa

    An open-source machine learning framework for building contextual AI assistants and agents that automate text and voice-based conversations.

  • Project
    agilexrobotics/piper_sdk

    A Python SDK for the Piper robot arm that processes CAN bus data frames into custom data types for easier hardware control and monitoring.