RasaGPT

Free

RasaGPT is the first headless LLM chatbot platform built on top of Rasa and Langchain. ![GitHub Repo stars](https://img.shields.io/github/stars/paulpierre/RasaGPT?style=social)

FreeFree tier

Inputs: textOutputs: text

Type

Open Source

Visit Website GitHub (40 stars)

About RasaGPT

RasaGPT is the first headless LLM chatbot platform built on top of Rasa and Langchain. It serves as boilerplate and a reference implementation for integrating Rasa with an LLM library like Langchain for indexing, retrieval, and context injection. Built with Rasa, FastAPI, Langchain, LlamaIndex, SQLModel, pgvector, ngrok, and Telegram, it offers a full-stack API that allows an LLM to learn from an arbitrary corpus of data. Features include document upload and automatic re-training via FastAPI, document versioning, customizable async endpoints and database models, human handoff detection, automatic tag generation, multi-tenancy, and Dockerized support for macOS. The project is open source and free, but the author explicitly warns it is not production code and contains security vulnerabilities.

Key Features

Headless LLM chatbot platform integrating Rasa and Langchain

Full application and API with FastAPI backend

Upload documents and automatically train the LLM on arbitrary corpora

Document versioning and automatic re-training on upload

Customizable async endpoints and database models via SQLModel

Bot determines when human handoff is necessary

Automatic generation of tags based on user questions and responses

Multi-tenancy support (not native in Rasa)

Reverse proxy support via ngrok

Dockerized deployment for macOS (with instructions for Linux/Windows)

Pros & Cons

Pros

Open source and free to use
Comprehensive boilerplate saving significant implementation time
Supports document upload and automatic re-training
Multi-tenancy support out of the box
Dockerized for easy setup on macOS
Integrates popular tools: Rasa, Langchain, FastAPI, pgvector, Telegram
Includes human handoff logic and automatic tag generation

Cons

Explicitly not production-ready; contains prompt injection and security vulnerabilities
Docker images are specific to macOS (aarch64) by default, requiring adjustments for Linux/Windows
Limited documentation beyond the README and a separate resource site
May require significant customization for real-world production use

Best For

Building custom chatbots with LLM-powered responsesIntegrating Rasa with Langchain for context injection and retrievalCreating Telegram bots that leverage document-based Q&APrototyping LLM-powered conversational agents with a full-stack APILearning how to combine Rasa, FastAPI, and vector databases

FAQ

What is RasaGPT?

RasaGPT is the first headless LLM chatbot platform built on top of Rasa and Langchain. It is a boilerplate and reference implementation for integrating Rasa with an LLM library like Langchain for indexing, retrieval, and context injection, including a Telegram bot.

Is RasaGPT production-ready?

No. The author explicitly warns that this is far from production code and is rife with prompt injection and general security vulnerabilities. It is intended as a reference implementation and starting point.

What technologies does RasaGPT use?

It uses Rasa, FastAPI, Langchain, LlamaIndex, SQLModel, pgvector, ngrok, and Telegram. It runs in Docker on macOS and supports file upload for training.

How do I get started with RasaGPT?

Clone the repository, copy the .env-example to .env and fill in credentials, then run `make install` to build and start the Docker containers.