OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Chat & Assistants

Rasagpt

Rasagpt is a Python-based, self-hosted chatbot reference implementation that connects Rasa and Telegram with LLM libraries for document indexing, retrieval and context injection. Its stack includes FastAPI, SQLModel, PostgreSQL with pgvector, Docker and ngrok.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Rasagpt — OpenViST project profile in Chat & Assistants.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want a self-hosted starting point for connecting a Rasa chatbot to LLM-based retrieval, especially for a Telegram prototype backed by indexed content.

Who it is for

  • Developers prototyping a Rasa chatbot with LLM-powered retrieval.
  • Teams building a Telegram chatbot that answers from an indexed knowledge corpus.
  • Engineers adapting chatbot APIs, data models or session and multi-tenant handling.

Documented uses

  • Create a chatbot that retrieves information from uploaded or indexed documents and injects relevant context into LLM responses.
  • Use Telegram as the chat channel in a Rasa-based reference implementation.
  • Adapt API endpoints and database models using FastAPI and SQLModel.
Before you choose

Requirements & limits

The repository is public and MIT-licensed. The documented setup requires local development tools and external API, Telegram bot and ngrok credentials.

  • Rasa and Telegram are the reference implementation's chatbot framework and channel.
  • LangChain or LlamaIndex provide the described indexing and retrieval layer; documentation does not establish support for every model or data source.
  • The README documents Docker Compose and local API development setup paths.
  • Linux and Windows users are told to change the Rasa image name in two repository files.
  • The project explicitly cautions that it is not production-ready and has prompt-injection and general security vulnerabilities.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare the repository and install with Docker Compose

Set up Rasagpt using the repository's documented Docker-oriented Makefile path.

Before you start

  • Python 3.9
  • Docker and Docker Compose
  • Make
  • SQLModel
  • OpenAI API key

Steps

  1. Clone the Rasagpt repository and change into its directory.

  2. Copy .env-example to .env and add the required credentials.

  3. Run the command below from the repository root.

    Documented command
    make install

Expected outcome

The README describes the Makefile as installing and running the services in order; it also identifies `make run` for subsequent runs.

Guide limitations

  • The README warns that Linux and Windows users need to change the Rasa image name in the Docker Compose file and actions Dockerfile.
  • The project warns that it is not production-ready and has prompt-injection and general security vulnerabilities.
Documentation 1 ↗
Source-based practical guide

Set up the local API development environment

Prepare the API component using its documented local Python environment path.

Before you start

  • Python 3.9
  • Make
  • SQLModel
  • OpenAI API key
  • Telegram bot credentials

Steps

  1. Clone the Rasagpt repository and change into its directory.

  2. Copy .env-example to .env and add the required credentials.

  3. Change into app/api and run the command below.

    Documented command
    make install

Expected outcome

The README describes this path as creating a local virtual environment for API development; it lists `make run` for running it again after installation.

Guide limitations

  • This path focuses on development on top of the API rather than installing all services through Docker Compose.
  • The project warns that it is not production-ready and has prompt-injection and general security vulnerabilities.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
11 October 2026
Original website title
paulpierre/RasaGPT · Official homepage title · retained in its source language
How we review sourcesSuggest a correction