OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Autonomous Agentic Rag

Autonomous Agentic Rag is an MIT-licensed Jupyter Notebook project describing a self-improving agentic RAG pipeline. Specialist agents generate outputs using standard operating procedures, evaluators score results across several goals, and diagnostic and SOP-architect agents propose revisions for further evaluation. The example targets healthcare and clinical-trial design, combining unstructured sources with structur

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Autonomous Agentic Rag — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use the project as a notebook-based reference when investigating how evaluation-driven changes to prompts, agent coordination or retrieval procedures could be structured in a RAG workflow.

Who it is for

  • Developers exploring agentic RAG and iterative retrieval-strategy optimization.
  • Technical readers prototyping multi-agent workflows for healthcare or clinical-trial design.

Documented uses

  • Model a workflow in which specialist agents produce a multi-source document from a high-level concept.
  • Evaluate outputs across goals such as accuracy, feasibility and compliance.
  • Use evaluation results to diagnose weaknesses and propose revised SOPs.
  • Compare candidate procedures by their trade-offs and present options for human selection.
Before you choose

Requirements & limits

The public GitHub repository is identified as open source under the MIT license.

  • The example separates unstructured sources, indexed with FAISS, from structured clinical data stored with DuckDB.
  • The documented stack includes LangChain, LangGraph, Ollama and LangSmith, as well as data-processing and embedding libraries.
  • The setup uses local model clients and identifies LangSmith API-key and Entrez-email environment variables as requirements.
  • The described evaluation loop identifies Pareto trade-offs for human decision-making rather than removing human review.

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 notebook environment and load credentials

Install the listed Python libraries and load LangSmith and Entrez credentials from a .env file.

Before you start

  • A Python notebook environment supporting the documented %pip install command.
  • A .env file containing LANGCHAIN_API_KEY and ENTREZ_EMAIL.

Steps

  1. Run the notebook's documented %pip install command to install the listed libraries.

  2. Create a .env file with LANGCHAIN_API_KEY and ENTREZ_EMAIL; do not commit the file to version control.

  3. In the notebook, call load_dotenv() and check that both variables are present in os.environ.

  4. Set LANGCHAIN_PROJECT to AI_Clinical_Trials_Architect.

Expected outcome

The listed dependencies are installed and the notebook loads the required credentials into its environment, with a LangSmith project name configured.

Guide limitations

  • This covers environment and credential setup only; it does not document a complete end-to-end run.
  • The source warns against committing the .env file. It does not specify an operating system or Python version.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
8 October 2026
Original website title
FareedKhan-dev/autonomous-agentic-rag · Official homepage title · retained in its source language
How we review sourcesSuggest a correction