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

Mercury

Mercury is an open-source TypeScript web app template for chatting with uploaded documents and selected websites. It uses text extraction, chunking, OpenAI embeddings, Pinecone similarity search, and GPT-3 responses to support conversations with history and source citations.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Mercury — 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 Mercury when you need a web-app starting point for conversational search over a defined set of files or selected webpages, and its documented OpenAI and Pinecone stack fits your implementation.

Who it is for

  • Developers building a document-question-answering web app.
  • Teams prototyping a conversational assistant grounded in selected web pages.
  • Builders who want a documented OpenAI-and-Pinecone retrieval workflow to adapt.

Documented uses

  • Ask questions about uploaded PDF, DOCX, Markdown, TXT, PNG, JPG, HTML, or JSON files.
  • Ground conversations in selected webpages scraped and indexed by the app.
  • Continue a discussion using chat history and review cited source material.
  • Use an existing file-ingestion and retrieval pattern as a starting point for a custom knowledge chat.
Before you choose

Requirements & limits

  • The documented file list includes PDF, DOCX, Markdown, TXT, PNG, JPG, HTML, and JSON; CSV and PPTX are listed as coming soon.
  • File and webpage text is split into 1,000-character chunks before embedding.
  • The documented retrieval flow uses OpenAI's text-embedding-ada-002 embeddings, Pinecone, and GPT-3.
  • Local setup requires OpenAI and Pinecone credentials and a Pinecone index with 1,536 dimensions.
  • The repository is public and its listed license is Unlicense.

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 Mercury for local development

Set up the documented app template and configure its OpenAI and Pinecone credentials.

Before you start

  • Node.js/npm environment for installing dependencies
  • OpenAI API key
  • Pinecone API key, environment name, and index name
  • Pinecone index configured with 1,536 dimensions

Steps

  1. Clone the repository or create a project from it with the documented degit command, then install dependencies with npm i.

  2. Create a Pinecone index with 1,536 dimensions and note its API key, environment name, and index name.

  3. Copy .env.example to .env.local and fill in OPENAI_API_KEY, PINECONE_API_KEY, PINECONE_ENVIRONMENT, and PINECONE_INDEX_NAME.

  4. Start the local app with npm run dev and open http://localhost:3000.

Expected outcome

A locally started Mercury app configured to use the documented OpenAI and Pinecone services.

Guide limitations

  • This setup recipe does not establish that file ingestion or chat execution has been tested.
  • The README marks CSV, PPTX, and Notion support as coming soon.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
Jordan-Gilliam/ai-template · Official homepage title · retained in its source language
How we review sourcesSuggest a correction