Skip to content

In today's digital landscape, misinformation can spread rapidly, making it crucial to validate online sources. This repository aims to empower users with an AI Agents-powered tool to assess the trustworthiness of information found on the web.

License

Notifications You must be signed in to change notification settings

phamkinhquoc2002/multi-agents-validator

Repository files navigation

Ragification Agents 🤖

An experiment using autogen AI Conversation Framework to create a team of AI Agents to assess quality and trustworthiness of any types of content from the internet.

Multi-AI Agent Conversation Framework

Techstack used: streamlit, langchain, autogen.

Description: the system contains two agents: researcher and questioner. The first one is programmed to extract key insights from the loaded documents, the second one uses those insights as an input to perform a validation framework, using a search_tool (Tavily) and ReAct prompt engineering.


framework


Run it locally

streamlit run app.py

About

In today's digital landscape, misinformation can spread rapidly, making it crucial to validate online sources. This repository aims to empower users with an AI Agents-powered tool to assess the trustworthiness of information found on the web.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published