OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Rag App On Aws

Rag App On Aws is an MIT-licensed Python project that provisions an AWS retrieval-augmented generation application with Terraform. It processes uploaded documents into embeddings stored in PostgreSQL with pgvector, answers queries using Google Gemini models, and provides a Streamlit interface for login, uploads, queries and evaluation. A separate remote MCP server connects the workflow to SerpAPI web search.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Rag App On Aws — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want a project starting point for deploying a document-querying RAG application on AWS and can configure its cloud resources and model integrations.

Who it is for

  • Developers building document-question-answering applications on AWS.
  • Teams seeking a Terraform-managed backend and a Streamlit interface for document uploads, queries and evaluation.

Documented uses

  • Provision AWS infrastructure for a RAG application.
  • Upload documents, process them into embeddings and query their contents.
  • Add real-time web search to a query workflow through the included remote MCP server.
  • Use a Streamlit frontend for login, uploads, querying and evaluation.
Before you choose

Requirements & limits

The repository is public and specifies an MIT license. Running the infrastructure can incur AWS charges; the README identifies RDS and NAT Gateway as potential costs when active.

  • Deployment provisions AWS services, including Lambda, API Gateway, S3, RDS and Cognito.
  • Gemini models handle embeddings and answer generation.
  • The web-search MCP server must be reachable by AWS Lambda; the README describes exposing a locally run server through a tunnel.
  • The repository describes GitHub Actions deployment and cleanup workflows and dev, staging and production environment configurations.

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

Configure and start the local web-search MCP server

Run the repository’s SerpAPI-backed MCP server locally and expose it for remote access during testing.

Before you start

  • A clone of the repository, including its mcp_servers directory.
  • A SerpAPI API key.
  • Windows PowerShell for the documented Cloudflare Tunnel option, or Windows PowerShell/Git Bash for the Serveo option.

Steps

  1. Create a .env file containing SERPAPI_API_KEY=your_serpai_api_key.

  2. Start the server with the command below.

    Documented command
    python web_search_mcp_server.py --host localhost --port 8000

  3. Expose the local server using either the documented Cloudflare Tunnel commands or the Serveo SSH tunnel command.

  4. Use the resulting public URL for external access to the MCP server in the RAG UI.

Expected outcome

The MCP server runs locally on port 8000 and can be reached through a public tunnel for testing.

Guide limitations

  • AWS Lambda cannot access localhost, so remote access requires exposing the local server through a tunnel.
  • The guide documents this for testing; it does not establish production deployment or security properties of the tunnel.
Documentation 1 ↗
Source-based practical guide

Set up the Streamlit UI locally after AWS resources are deployed

Install and configure the local Streamlit frontend to connect to the deployed RAG application.

Before you start

  • A cloned repository with the rag_ui directory.
  • AWS resources deployed through the GitHub Actions pipeline.
  • The API Gateway URL and Cognito client ID for the deployment.
  • Python and the ability to install and run the documented uv commands.

Steps

  1. From the cloned repository, navigate to rag-app-on-aws/rag_ui.

  2. Install uv if it is not already installed, create a virtual environment with the command below, and activate it using the documented platform-specific command.

    Documented command
    uv venv

  3. Install the UI dependencies with the command below.

    Documented command
    uv pip install -r requirements.txt

  4. Create a .env file setting API_ENDPOINT, UPLOAD_ENDPOINT=/upload, QUERY_ENDPOINT=/query, AUTH_ENDPOINT=/auth, DEFAULT_USER_ID=test-user, COGNITO_CLIENT_ID, and ENABLE_EVALUATION="true".

  5. After the GitHub Actions pipeline completes successfully, use the environment variables file provided by the pipeline as described in the UI setup instructions.

Expected outcome

The UI environment and configuration are prepared to interact locally with the deployed backend.

Guide limitations

  • The documented setup depends on AWS resources having already been deployed; these steps do not deploy the backend or specify a command to launch the UI.
  • The API URL and Cognito client ID must be supplied for the target deployment.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for Retrieval-augmented applications

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
Rag App On AwsProvision AWS infrastructure for a RAG application.Cloud, WebThe repository is public and specifies an MIT license. Running the infrastructure can incur AWS charges; the README identifies RDS and NAT Gateway as potential costs when active.
Rag Powered AI Assistant Transforming Data Retrieval And Analysis Across The Web And PdfsRetrieve relevant passages from web and PDF content to answer a question.Web
LangChain in Supabase Edge FunctionsInitialize and start a Supabase project locally.Cloud, Server
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
genieincodebottle/rag-app-on-aws · Official homepage title · retained in its source language
How we review sourcesSuggest a correction