https://github.com/drewbloom/multiAgentWorkflow

Testing multiagent AI flows for CRUD, search, and document construction from databases
https://github.com/drewbloom/multiAgentWorkflow

Last synced: 3 days ago
JSON representation

Testing multiagent AI flows for CRUD, search, and document construction from databases

• Host: GitHub
• URL: https://github.com/drewbloom/multiAgentWorkflow
• Owner: drewbloom
• Created: 2024-11-10T20:06:11.000Z (about 2 months ago)
• Default Branch: main
• Last Pushed: 2024-11-26T22:14:57.000Z (about 1 month ago)
• Last Synced: 2024-11-26T23:21:56.558Z (about 1 month ago)
• Language: Python
• Size: 646 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome_ai_agents - Multiagentworkflow - Testing multiagent AI flows for CRUD, search, and document construction from databases (Building / Testing)
• awesome_ai_agents - Multiagentworkflow - Testing multiagent AI flows for CRUD, search, and document construction from databases (Building / Testing)

README

        
### Multi-Agent Workflow Demo

Welcome to an in-progress demo for a multi-agent setup.

## Current Status

- Streamlit app has authentication and can handle both text and speech modalities

- AI assistant has tools to understand and query the attached database to help the user

# Next steps:

- Save conversation memory and cache database knowledge for each user

- Integrate with SharePoint data environment to expand assistant capabilities

- Create a vector store of the database as an additional tool for quick retrieval

    - This will be necessary for larger dbs, or dbs containing large documents

## Overall Goals

- Offer a user both text and speech modalities to accomplish tasks utilizing their own data

- Use common Python dependencies to connect together multiple AI agents that work together to determine how best to serve the user

- Include tool use that allows those agents to engage in semantic search and make SQL queries

- Integrate with sample data sources so this model can be replicated easily with other databases using SQL

## Replicating This Setup

- Must use a SQLite database or db file

- app folder contains deployments

- agents folder contains the agent setups

- database folder contains the database and scripting for interacting with it

* This may be all be migrated into utils.py once an appropriate setup is determined